Toggle Poster Visibility
Remarks
Mon May 03 12:00 AM -- 01:00 AM (PDT)
Opening Remarks
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Deciphering and Optimizing Multi-Task Learning: a Random Matrix Approach
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Uncertainty Estimation and Calibration with Finite-State Probabilistic RNNs
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Overfitting for Fun and Profit: Instance-Adaptive Data Compression
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Is Label Smoothing Truly Incompatible with Knowledge Distillation: An Empirical Study
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Set Prediction without Imposing Structure as Conditional Density Estimation
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
What Makes Instance Discrimination Good for Transfer Learning?
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Batch Reinforcement Learning Through Continuation Method
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
A Good Image Generator Is What You Need for High-Resolution Video Synthesis
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Training with Quantization Noise for Extreme Model Compression
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Domain Generalization with MixStyle
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
On InstaHide, Phase Retrieval, and Sparse Matrix Factorization
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Gauge Equivariant Mesh CNNs: Anisotropic convolutions on geometric graphs
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Mutual Information State Intrinsic Control
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
On the Transfer of Disentangled Representations in Realistic Settings
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
LEAF: A Learnable Frontend for Audio Classification
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Tomographic Auto-Encoder: Unsupervised Bayesian Recovery of Corrupted Data
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Wasserstein-2 Generative Networks
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
MELR: Meta-Learning via Modeling Episode-Level Relationships for Few-Shot Learning
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Meta-GMVAE: Mixture of Gaussian VAE for Unsupervised Meta-Learning
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bits
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
PSTNet: Point Spatio-Temporal Convolution on Point Cloud Sequences
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
The Unreasonable Effectiveness of Patches in Deep Convolutional Kernels Methods
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Improve Object Detection with Feature-based Knowledge Distillation: Towards Accurate and Efficient Detectors
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Noise against noise: stochastic label noise helps combat inherent label noise
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
MODALS: Modality-agnostic Automated Data Augmentation in the Latent Space
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Signatory: differentiable computations of the signature and logsignature transforms, on both CPU and GPU
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
WaNet - Imperceptible Warping-based Backdoor Attack
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
On the Universality of the Double Descent Peak in Ridgeless Regression
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
FairFil: Contrastive Neural Debiasing Method for Pretrained Text Encoders
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Scalable Transfer Learning with Expert Models
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Neural Approximate Sufficient Statistics for Implicit Models
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
MetaNorm: Learning to Normalize Few-Shot Batches Across Domains
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularization
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Towards Robust Neural Networks via Close-loop Control
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Exploring Balanced Feature Spaces for Representation Learning
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Randomized Ensembled Double Q-Learning: Learning Fast Without a Model
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Rethinking the Role of Gradient-based Attribution Methods for Model Interpretability
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Wasserstein Embedding for Graph Learning
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Isometric Transformation Invariant and Equivariant Graph Convolutional Networks
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Learning N:M Fine-grained Structured Sparse Neural Networks From Scratch
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Towards Robustness Against Natural Language Word Substitutions
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
On Learning Universal Representations Across Languages
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Neural Jump Ordinary Differential Equations: Consistent Continuous-Time Prediction and Filtering
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Revisiting Locally Supervised Learning: an Alternative to End-to-end Training
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Predicting Infectiousness for Proactive Contact Tracing
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Progressive Skeletonization: Trimming more fat from a network at initialization
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Implicit Normalizing Flows
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
SALD: Sign Agnostic Learning with Derivatives
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
Does enhanced shape bias improve neural network robustness to common corruptions?
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Interpreting and Boosting Dropout from a Game-Theoretic View
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
ResNet After All: Neural ODEs and Their Numerical Solution
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT) @ Virtual
QPLEX: Duplex Dueling Multi-Agent Q-Learning
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
A Unified Approach to Interpreting and Boosting Adversarial Transferability
[ Paper ]
Poster
Mon May 03 01:00 AM -- 03:00 AM (PDT)
Solving Compositional Reinforcement Learning Problems via Task Reduction
[ Paper ]
Oral
Mon May 03 03:15 AM -- 03:30 AM (PDT)
Free Lunch for Few-shot Learning: Distribution Calibration
[ Paper ]
Spotlight
Mon May 03 03:30 AM -- 03:40 AM (PDT)
Deciphering and Optimizing Multi-Task Learning: a Random Matrix Approach
[ Paper ]
Spotlight
Mon May 03 03:40 AM -- 03:50 AM (PDT)
Generalization in data-driven models of primary visual cortex
[ Paper ]
Oral
Mon May 03 04:00 AM -- 04:15 AM (PDT)
Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding
[ Paper ]
Oral
Mon May 03 04:15 AM -- 04:30 AM (PDT)
A Distributional Approach to Controlled Text Generation
[ Paper ]
Spotlight
Mon May 03 04:30 AM -- 04:40 AM (PDT)
The Intrinsic Dimension of Images and Its Impact on Learning
[ Paper ]
Oral
Mon May 03 05:00 AM -- 05:15 AM (PDT)
Geometry-aware Instance-reweighted Adversarial Training
[ Paper ]
Oral
Mon May 03 05:15 AM -- 05:30 AM (PDT)
Do 2D GANs Know 3D Shape? Unsupervised 3D Shape Reconstruction from 2D Image GANs
[ Paper ]
Oral
Mon May 03 05:30 AM -- 05:45 AM (PDT)
Rethinking the Role of Gradient-based Attribution Methods for Model Interpretability
[ Paper ]
Spotlight
Mon May 03 05:45 AM -- 05:55 AM (PDT)
Contrastive Divergence Learning is a Time Reversal Adversarial Game
[ Paper ]
Break
Mon May 03 06:15 AM -- 08:00 AM (PDT)
BREAK:
Invited Talk
Mon May 03 08:00 AM -- 09:00 AM (PDT) @ Virtual
Moving beyond the fairness rhetoric in machine learning
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Structured Prediction as Translation between Augmented Natural Languages
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Into the Wild with AudioScope: Unsupervised Audio-Visual Separation of On-Screen Sounds
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Trajectory Prediction using Equivariant Continuous Convolution
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
On the role of planning in model-based deep reinforcement learning
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Shapley Explanation Networks
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Fast convergence of stochastic subgradient method under interpolation
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
The role of Disentanglement in Generalisation
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Learning-based Support Estimation in Sublinear Time
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Accelerating Convergence of Replica Exchange Stochastic Gradient MCMC via Variance Reduction
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Multi-Time Attention Networks for Irregularly Sampled Time Series
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
InfoBERT: Improving Robustness of Language Models from An Information Theoretic Perspective
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
X2T: Training an X-to-Text Typing Interface with Online Learning from User Feedback
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Unsupervised Meta-Learning through Latent-Space Interpolation in Generative Models
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Sparse encoding for more-interpretable feature-selecting representations in probabilistic matrix factorization
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
WrapNet: Neural Net Inference with Ultra-Low-Precision Arithmetic
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
MultiModalQA: complex question answering over text, tables and images
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Scaling Symbolic Methods using Gradients for Neural Model Explanation
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
What Can You Learn From Your Muscles? Learning Visual Representation from Human Interactions
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Predicting Classification Accuracy When Adding New Unobserved Classes
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Understanding the failure modes of out-of-distribution generalization
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Revisiting Few-sample BERT Fine-tuning
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Overparameterisation and worst-case generalisation: friend or foe?
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Rethinking Embedding Coupling in Pre-trained Language Models
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
The Risks of Invariant Risk Minimization
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Self-Supervised Policy Adaptation during Deployment
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Learning with AMIGo: Adversarially Motivated Intrinsic Goals
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
On the Impossibility of Global Convergence in Multi-Loss Optimization
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Selectivity considered harmful: evaluating the causal impact of class selectivity in DNNs
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
What Should Not Be Contrastive in Contrastive Learning
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Off-Dynamics Reinforcement Learning: Training for Transfer with Domain Classifiers
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Language-Agnostic Representation Learning of Source Code from Structure and Context
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Training GANs with Stronger Augmentations via Contrastive Discriminator
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Graph Convolution with Low-rank Learnable Local Filters
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Learning "What-if" Explanations for Sequential Decision-Making
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Learning Invariant Representations for Reinforcement Learning without Reconstruction
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Shape-Texture Debiased Neural Network Training
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Using latent space regression to analyze and leverage compositionality in GANs
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
A statistical theory of cold posteriors in deep neural networks
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Parameter Efficient Multimodal Transformers for Video Representation Learning
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Seq2Tens: An Efficient Representation of Sequences by Low-Rank Tensor Projections
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Effective Distributed Learning with Random Features: Improved Bounds and Algorithms
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Reset-Free Lifelong Learning with Skill-Space Planning
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
GraPPa: Grammar-Augmented Pre-Training for Table Semantic Parsing
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Disentangling 3D Prototypical Networks for Few-Shot Concept Learning
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Grounding Physical Concepts of Objects and Events Through Dynamic Visual Reasoning
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
LambdaNetworks: Modeling long-range Interactions without Attention
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
The Traveling Observer Model: Multi-task Learning Through Spatial Variable Embeddings
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Gradient Projection Memory for Continual Learning
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Universal approximation power of deep residual neural networks via nonlinear control theory
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
On the Universality of Rotation Equivariant Point Cloud Networks
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Plan-Based Relaxed Reward Shaping for Goal-Directed Tasks
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Single-Photon Image Classification
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Saliency is a Possible Red Herring When Diagnosing Poor Generalization
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Symmetry-Aware Actor-Critic for 3D Molecular Design
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
On Statistical Bias In Active Learning: How and When to Fix It
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Planning from Pixels using Inverse Dynamics Models
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
PAC Confidence Predictions for Deep Neural Network Classifiers
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
MoVie: Revisiting Modulated Convolutions for Visual Counting and Beyond
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
NeMo: Neural Mesh Models of Contrastive Features for Robust 3D Pose Estimation
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Vector-output ReLU Neural Network Problems are Copositive Programs: Convex Analysis of Two Layer Networks and Polynomial-time Algorithms
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Learning from others' mistakes: Avoiding dataset biases without modeling them
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Predicting Inductive Biases of Pre-Trained Models
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Uncertainty Sets for Image Classifiers using Conformal Prediction
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Conditional Negative Sampling for Contrastive Learning of Visual Representations
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Learning Hyperbolic Representations of Topological Features
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Extracting Strong Policies for Robotics Tasks from Zero-Order Trajectory Optimizers
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Learning Structural Edits via Incremental Tree Transformations
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Representation learning for improved interpretability and classification accuracy of clinical factors from EEG
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Multi-Level Local SGD: Distributed SGD for Heterogeneous Hierarchical Networks
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT)
Pruning Neural Networks at Initialization: Why Are We Missing the Mark?
[ Paper ]
Poster
Mon May 03 09:00 AM -- 11:00 AM (PDT) @ Virtual
Share or Not? Learning to Schedule Language-Specific Capacity for Multilingual Translation
Oral
Mon May 03 11:00 AM -- 11:15 AM (PDT)
Federated Learning Based on Dynamic Regularization
[ Paper ]
Oral
Mon May 03 11:15 AM -- 11:30 AM (PDT)
Gradient Projection Memory for Continual Learning
[ Paper ]
Oral
Mon May 03 11:30 AM -- 11:45 AM (PDT)
Growing Efficient Deep Networks by Structured Continuous Sparsification
[ Paper ]
Spotlight
Mon May 03 11:45 AM -- 11:55 AM (PDT)
Geometry-Aware Gradient Algorithms for Neural Architecture Search
[ Paper ]
Spotlight
Mon May 03 12:15 PM -- 12:25 PM (PDT)
On the Theory of Implicit Deep Learning: Global Convergence with Implicit Layers
[ Paper ]
Spotlight
Mon May 03 12:25 PM -- 12:35 PM (PDT)
Sharpness-aware Minimization for Efficiently Improving Generalization
[ Paper ]
Spotlight
Mon May 03 12:35 PM -- 12:45 PM (PDT)
Systematic generalisation with group invariant predictions
[ Paper ]
Spotlight
Mon May 03 12:45 PM -- 12:55 PM (PDT)
On Statistical Bias In Active Learning: How and When to Fix It
[ Paper ]
Spotlight
Mon May 03 12:55 PM -- 01:05 PM (PDT)
Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images
[ Paper ]
Spotlight
Mon May 03 01:20 PM -- 01:30 PM (PDT)
Uncertainty Sets for Image Classifiers using Conformal Prediction
[ Paper ]
Spotlight
Mon May 03 01:30 PM -- 01:40 PM (PDT)
PMI-Masking: Principled masking of correlated spans
[ Paper ]
Spotlight
Mon May 03 01:40 PM -- 01:50 PM (PDT)
Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Models
[ Paper ]
Spotlight
Mon May 03 01:50 PM -- 02:00 PM (PDT)
Watch-And-Help: A Challenge for Social Perception and Human-AI Collaboration
[ Paper ]
Spotlight
Mon May 03 02:00 PM -- 02:10 PM (PDT)
Predicting Infectiousness for Proactive Contact Tracing
[ Paper ]
Break
Mon May 03 02:30 PM -- 04:00 PM (PDT)
BREAK:
Invited Talk
Mon May 03 04:00 PM -- 05:00 PM (PDT) @ Virtual
Commonsense AI: Myth and Truth
[
Slides]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Self-training For Few-shot Transfer Across Extreme Task Differences
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Benefit of deep learning with non-convex noisy gradient descent: Provable excess risk bound and superiority to kernel methods
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Improved Estimation of Concentration Under $\ell_p$-Norm Distance Metrics Using Half Spaces
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
VA-RED$^2$: Video Adaptive Redundancy Reduction
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Federated Learning Based on Dynamic Regularization
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
When Optimizing $f$-Divergence is Robust with Label Noise
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
The Role of Momentum Parameters in the Optimal Convergence of Adaptive Polyak's Heavy-ball Methods
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
SenSeI: Sensitive Set Invariance for Enforcing Individual Fairness
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Why resampling outperforms reweighting for correcting sampling bias with stochastic gradients
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Incorporating Symmetry into Deep Dynamics Models for Improved Generalization
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Rethinking Architecture Selection in Differentiable NAS
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Score-Based Generative Modeling through Stochastic Differential Equations
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
UPDeT: Universal Multi-agent RL via Policy Decoupling with Transformers
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Representation Learning for Sequence Data with Deep Autoencoding Predictive Components
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
What are the Statistical Limits of Offline RL with Linear Function Approximation?
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Model Patching: Closing the Subgroup Performance Gap with Data Augmentation
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Sequential Density Ratio Estimation for Simultaneous Optimization of Speed and Accuracy
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Parrot: Data-Driven Behavioral Priors for Reinforcement Learning
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
On the geometry of generalization and memorization in deep neural networks
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
When does preconditioning help or hurt generalization?
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Layer-adaptive Sparsity for the Magnitude-based Pruning
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Explaining the Efficacy of Counterfactually Augmented Data
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Contextual Transformation Networks for Online Continual Learning
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Online Adversarial Purification based on Self-supervised Learning
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Learning Energy-Based Models by Diffusion Recovery Likelihood
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
On Fast Adversarial Robustness Adaptation in Model-Agnostic Meta-Learning
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Robust and Generalizable Visual Representation Learning via Random Convolutions
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
One Network Fits All? Modular versus Monolithic Task Formulations in Neural Networks
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
On Dyadic Fairness: Exploring and Mitigating Bias in Graph Connections
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Offline Model-Based Optimization via Normalized Maximum Likelihood Estimation
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Regularization Matters in Policy Optimization - An Empirical Study on Continuous Control
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Regularized Inverse Reinforcement Learning
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Unlearnable Examples: Making Personal Data Unexploitable
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Undistillable: Making A Nasty Teacher That CANNOT teach students
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Decentralized Attribution of Generative Models
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
MALI: A memory efficient and reverse accurate integrator for Neural ODEs
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Personalized Federated Learning with First Order Model Optimization
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Remembering for the Right Reasons: Explanations Reduce Catastrophic Forgetting
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
The Intrinsic Dimension of Images and Its Impact on Learning
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Rethinking Positional Encoding in Language Pre-training
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Robust Curriculum Learning: from clean label detection to noisy label self-correction
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Deep Partition Aggregation: Provable Defenses against General Poisoning Attacks
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
SOLAR: Sparse Orthogonal Learned and Random Embeddings
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
WaveGrad: Estimating Gradients for Waveform Generation
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Latent Skill Planning for Exploration and Transfer
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Long Live the Lottery: The Existence of Winning Tickets in Lifelong Learning
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
SAFENet: A Secure, Accurate and Fast Neural Network Inference
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
PseudoSeg: Designing Pseudo Labels for Semantic Segmentation
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
MONGOOSE: A Learnable LSH Framework for Efficient Neural Network Training
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Learning a Latent Simplex in Input Sparsity Time
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
MixKD: Towards Efficient Distillation of Large-scale Language Models
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Optimal Regularization can Mitigate Double Descent
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Deberta: Decoding-Enhanced Bert With Disentangled Attention
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
MoPro: Webly Supervised Learning with Momentum Prototypes
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
The Deep Bootstrap Framework: Good Online Learners are Good Offline Generalizers
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
SCoRe: Pre-Training for Context Representation in Conversational Semantic Parsing
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Taking Notes on the Fly Helps Language Pre-Training
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Zero-shot Synthesis with Group-Supervised Learning
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Robust Reinforcement Learning on State Observations with Learned Optimal Adversary
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Optimizing Memory Placement using Evolutionary Graph Reinforcement Learning
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Selective Classification Can Magnify Disparities Across Groups
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Bypassing the Ambient Dimension: Private SGD with Gradient Subspace Identification
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT)
Proximal Gradient Descent-Ascent: Variable Convergence under KŁ Geometry
[ Paper ]
Poster
Mon May 03 05:00 PM -- 07:00 PM (PDT) @ Virtual
Random Feature Attention
Oral
Mon May 03 07:00 PM -- 07:15 PM (PDT)
SMiRL: Surprise Minimizing Reinforcement Learning in Unstable Environments
[ Paper ]
Oral
Mon May 03 07:15 PM -- 07:30 PM (PDT)
Contrastive Explanations for Reinforcement Learning via Embedded Self Predictions
[ Paper ]
Oral
Mon May 03 07:30 PM -- 07:45 PM (PDT)
Parrot: Data-Driven Behavioral Priors for Reinforcement Learning
[ Paper ]
Spotlight
Mon May 03 07:45 PM -- 07:55 PM (PDT)
Structured Prediction as Translation between Augmented Natural Languages
[ Paper ]
Spotlight
Mon May 03 07:55 PM -- 08:05 PM (PDT)
Mathematical Reasoning via Self-supervised Skip-tree Training
[ Paper ]
Spotlight
Mon May 03 08:18 PM -- 08:28 PM (PDT)
Improving Adversarial Robustness via Channel-wise Activation Suppressing
[ Paper ]
Spotlight
Mon May 03 08:28 PM -- 08:38 PM (PDT)
Fast Geometric Projections for Local Robustness Certification
[ Paper ]
Spotlight
Mon May 03 08:48 PM -- 08:58 PM (PDT)
Dataset Inference: Ownership Resolution in Machine Learning
[ Paper ]
Spotlight
Mon May 03 08:58 PM -- 09:08 PM (PDT)
HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark
[ Paper ]
Oral
Mon May 03 09:21 PM -- 09:36 PM (PDT)
How Neural Networks Extrapolate: From Feedforward to Graph Neural Networks
[ Paper ]
Spotlight
Mon May 03 09:36 PM -- 09:46 PM (PDT)
Graph Convolution with Low-rank Learnable Local Filters
[ Paper ]
Spotlight
Mon May 03 09:46 PM -- 09:56 PM (PDT)
The Traveling Observer Model: Multi-task Learning Through Spatial Variable Embeddings
[ Paper ]
Spotlight
Mon May 03 09:56 PM -- 10:06 PM (PDT)
Meta-GMVAE: Mixture of Gaussian VAE for Unsupervised Meta-Learning
[ Paper ]
Break
Mon May 03 10:30 PM -- 10:30 PM (PDT)
BREAK:
Invited Talk
Tue May 04 12:00 AM -- 01:00 AM (PDT) @ Virtual
Geometric Deep Learning: the Erlangen Programme of ML
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Monte-Carlo Planning and Learning with Language Action Value Estimates
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Effective Abstract Reasoning with Dual-Contrast Network
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Image GANs meet Differentiable Rendering for Inverse Graphics and Interpretable 3D Neural Rendering
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Generalized Multimodal ELBO
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Spatial Dependency Networks: Neural Layers for Improved Generative Image Modeling
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Large-width functional asymptotics for deep Gaussian neural networks
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Deep Repulsive Clustering of Ordered Data Based on Order-Identity Decomposition
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Distributed Momentum for Byzantine-resilient Stochastic Gradient Descent
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Learning Incompressible Fluid Dynamics from Scratch - Towards Fast, Differentiable Fluid Models that Generalize
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Auxiliary Learning by Implicit Differentiation
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Contemplating Real-World Object Classification
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Scaling the Convex Barrier with Active Sets
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Refining Deep Generative Models via Discriminator Gradient Flow
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
A Block Minifloat Representation for Training Deep Neural Networks
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
MiCE: Mixture of Contrastive Experts for Unsupervised Image Clustering
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Winning the L2RPN Challenge: Power Grid Management via Semi-Markov Afterstate Actor-Critic
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Coping with Label Shift via Distributionally Robust Optimisation
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Rao-Blackwellizing the Straight-Through Gumbel-Softmax Gradient Estimator
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Sample-Efficient Automated Deep Reinforcement Learning
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Probing BERT in Hyperbolic Spaces
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Computational Separation Between Convolutional and Fully-Connected Networks
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Prototypical Contrastive Learning of Unsupervised Representations
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
BOIL: Towards Representation Change for Few-shot Learning
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Policy-Driven Attack: Learning to Query for Hard-label Black-box Adversarial Examples
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Multiscale Score Matching for Out-of-Distribution Detection
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Learning Accurate Entropy Model with Global Reference for Image Compression
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Class Normalization for (Continual)? Generalized Zero-Shot Learning
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
FedMix: Approximation of Mixup under Mean Augmented Federated Learning
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
AdaGCN: Adaboosting Graph Convolutional Networks into Deep Models
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Learning Better Structured Representations Using Low-rank Adaptive Label Smoothing
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Bayesian Context Aggregation for Neural Processes
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
A Trainable Optimal Transport Embedding for Feature Aggregation and its Relationship to Attention
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Improving Transformation Invariance in Contrastive Representation Learning
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Group Equivariant Conditional Neural Processes
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Complex Query Answering with Neural Link Predictors
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Practical Real Time Recurrent Learning with a Sparse Approximation
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
A Universal Representation Transformer Layer for Few-Shot Image Classification
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Learning Subgoal Representations with Slow Dynamics
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Activation-level uncertainty in deep neural networks
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Generalization in data-driven models of primary visual cortex
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Do 2D GANs Know 3D Shape? Unsupervised 3D Shape Reconstruction from 2D Image GANs
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Model-based micro-data reinforcement learning: what are the crucial model properties and which model to choose?
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
A Distributional Approach to Controlled Text Generation
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Intraclass clustering: an implicit learning ability that regularizes DNNs
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
On Self-Supervised Image Representations for GAN Evaluation
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
IDF++: Analyzing and Improving Integer Discrete Flows for Lossless Compression
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
PDE-Driven Spatiotemporal Disentanglement
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Conformation-Guided Molecular Representation with Hamiltonian Neural Networks
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Lossless Compression of Structured Convolutional Models via Lifting
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Disambiguating Symbolic Expressions in Informal Documents
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Prediction and generalisation over directed actions by grid cells
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Exemplary Natural Images Explain CNN Activations Better than State-of-the-Art Feature Visualization
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Identifying nonlinear dynamical systems with multiple time scales and long-range dependencies
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
Accurate Learning of Graph Representations with Graph Multiset Pooling
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
SkipW: Resource Adaptable RNN with Strict Upper Computational Limit
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
not-MIWAE: Deep Generative Modelling with Missing not at Random Data
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Efficient Certified Defenses Against Patch Attacks on Image Classifiers
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
GraphCodeBERT: Pre-training Code Representations with Data Flow
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Group Equivariant Stand-Alone Self-Attention For Vision
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT)
What they do when in doubt: a study of inductive biases in seq2seq learners
[ Paper ]
Poster
Tue May 04 01:00 AM -- 03:00 AM (PDT) @ Virtual
Capturing Label Characteristics in VAEs
Spotlight
Tue May 04 03:25 AM -- 03:35 AM (PDT)
Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking
[ Paper ]
Spotlight
Tue May 04 03:35 AM -- 03:45 AM (PDT)
Expressive Power of Invariant and Equivariant Graph Neural Networks
[ Paper ]
Spotlight
Tue May 04 03:45 AM -- 03:55 AM (PDT)
Gauge Equivariant Mesh CNNs: Anisotropic convolutions on geometric graphs
[ Paper ]
Oral
Tue May 04 04:08 AM -- 04:23 AM (PDT)
Rao-Blackwellizing the Straight-Through Gumbel-Softmax Gradient Estimator
[ Paper ]
Oral
Tue May 04 04:23 AM -- 04:38 AM (PDT)
Scalable Learning and MAP Inference for Nonsymmetric Determinantal Point Processes
[ Paper ]
Spotlight
Tue May 04 04:38 AM -- 04:48 AM (PDT)
Multivariate Probabilistic Time Series Forecasting via Conditioned Normalizing Flows
[ Paper ]
Spotlight
Tue May 04 04:48 AM -- 04:58 AM (PDT)
Noise against noise: stochastic label noise helps combat inherent label noise
[ Paper ]
Spotlight
Tue May 04 05:08 AM -- 05:18 AM (PDT)
Mutual Information State Intrinsic Control
[ Paper ]
Spotlight
Tue May 04 05:18 AM -- 05:28 AM (PDT)
Learning Incompressible Fluid Dynamics from Scratch - Towards Fast, Differentiable Fluid Models that Generalize
[ Paper ]
Spotlight
Tue May 04 05:28 AM -- 05:38 AM (PDT)
Identifying nonlinear dynamical systems with multiple time scales and long-range dependencies
[ Paper ]
Break
Tue May 04 06:00 AM -- 08:00 AM (PDT)
BREAK:
Invited Talk
Tue May 04 08:00 AM -- 09:00 AM (PDT) @ Virtual
AI in Finance: Scope and Examples
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
UMEC: Unified model and embedding compression for efficient recommendation systems
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Statistical inference for individual fairness
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Representation Learning via Invariant Causal Mechanisms
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Quantifying Differences in Reward Functions
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Training BatchNorm and Only BatchNorm: On the Expressive Power of Random Features in CNNs
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Supervised Contrastive Learning for Pre-trained Language Model Fine-tuning
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Decoupling Global and Local Representations via Invertible Generative Flows
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Mapping the Timescale Organization of Neural Language Models
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Recurrent Independent Mechanisms
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Understanding Over-parameterization in Generative Adversarial Networks
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Tradeoffs in Data Augmentation: An Empirical Study
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Shape or Texture: Understanding Discriminative Features in CNNs
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
C-Learning: Horizon-Aware Cumulative Accessibility Estimation
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
DINO: A Conditional Energy-Based GAN for Domain Translation
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
On the Origin of Implicit Regularization in Stochastic Gradient Descent
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Transient Non-stationarity and Generalisation in Deep Reinforcement Learning
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Scalable Bayesian Inverse Reinforcement Learning
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Learning from Protein Structure with Geometric Vector Perceptrons
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
NOVAS: Non-convex Optimization via Adaptive Stochastic Search for End-to-end Learning and Control
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
SMiRL: Surprise Minimizing Reinforcement Learning in Unstable Environments
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
On the mapping between Hopfield networks and Restricted Boltzmann Machines
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Distance-Based Regularisation of Deep Networks for Fine-Tuning
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Ringing ReLUs: Harmonic Distortion Analysis of Nonlinear Feedforward Networks
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Characterizing signal propagation to close the performance gap in unnormalized ResNets
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Uncertainty-aware Active Learning for Optimal Bayesian Classifier
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Unsupervised Object Keypoint Learning using Local Spatial Predictability
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Transformer protein language models are unsupervised structure learners
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Uncertainty Estimation in Autoregressive Structured Prediction
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Learning a Latent Search Space for Routing Problems using Variational Autoencoders
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Support-set bottlenecks for video-text representation learning
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Reducing the Computational Cost of Deep Generative Models with Binary Neural Networks
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
On the Theory of Implicit Deep Learning: Global Convergence with Implicit Layers
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Self-Supervised Learning of Compressed Video Representations
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Anatomy of Catastrophic Forgetting: Hidden Representations and Task Semantics
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Robust Pruning at Initialization
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Unsupervised Representation Learning for Time Series with Temporal Neighborhood Coding
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
DC3: A learning method for optimization with hard constraints
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Direction Matters: On the Implicit Bias of Stochastic Gradient Descent with Moderate Learning Rate
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Data-Efficient Reinforcement Learning with Self-Predictive Representations
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Learning Neural Event Functions for Ordinary Differential Equations
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Getting a CLUE: A Method for Explaining Uncertainty Estimates
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Single-Timescale Actor-Critic Provably Finds Globally Optimal Policy
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Physics-aware, probabilistic model order reduction with guaranteed stability
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Large Associative Memory Problem in Neurobiology and Machine Learning
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
SSD: A Unified Framework for Self-Supervised Outlier Detection
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Rank the Episodes: A Simple Approach for Exploration in Procedurally-Generated Environments
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Vulnerability-Aware Poisoning Mechanism for Online RL with Unknown Dynamics
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Iterative Empirical Game Solving via Single Policy Best Response
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
FairBatch: Batch Selection for Model Fairness
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Systematic generalisation with group invariant predictions
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Are wider nets better given the same number of parameters?
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Meta-learning Symmetries by Reparameterization
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
The geometry of integration in text classification RNNs
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Discovering a set of policies for the worst case reward
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Learning Robust State Abstractions for Hidden-Parameter Block MDPs
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Learning Value Functions in Deep Policy Gradients using Residual Variance
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Provable Rich Observation Reinforcement Learning with Combinatorial Latent States
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Global optimality of softmax policy gradient with single hidden layer neural networks in the mean-field regime
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Towards Resolving the Implicit Bias of Gradient Descent for Matrix Factorization: Greedy Low-Rank Learning
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Sharper Generalization Bounds for Learning with Gradient-dominated Objective Functions
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
VTNet: Visual Transformer Network for Object Goal Navigation
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Tent: Fully Test-Time Adaptation by Entropy Minimization
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT) @ Virtual
Text Generation by Learning from Demonstrations
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
DialoGraph: Incorporating Interpretable Strategy-Graph Networks into Negotiation Dialogues
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Interpreting Knowledge Graph Relation Representation from Word Embeddings
[ Paper ]
Poster
Tue May 04 09:00 AM -- 11:00 AM (PDT)
Clairvoyance: A Pipeline Toolkit for Medical Time Series
[ Paper ]
Oral
Tue May 04 11:00 AM -- 11:15 AM (PDT)
Iterated learning for emergent systematicity in VQA
[ Paper ]
Oral
Tue May 04 11:15 AM -- 11:30 AM (PDT)
Learning Generalizable Visual Representations via Interactive Gameplay
[ Paper ]
Spotlight
Tue May 04 11:30 AM -- 11:40 AM (PDT)
How Does Mixup Help With Robustness and Generalization?
[ Paper ]
Oral
Tue May 04 12:15 PM -- 12:30 PM (PDT)
Image GANs meet Differentiable Rendering for Inverse Graphics and Interpretable 3D Neural Rendering
[ Paper ]
Spotlight
Tue May 04 12:30 PM -- 12:40 PM (PDT)
Mind the Pad -- CNNs Can Develop Blind Spots
[ Paper ]
Spotlight
Tue May 04 12:40 PM -- 12:50 PM (PDT)
Implicit Convex Regularizers of CNN Architectures: Convex Optimization of Two- and Three-Layer Networks in Polynomial Time
[ Paper ]
Spotlight
Tue May 04 12:50 PM -- 01:00 PM (PDT)
Learning from Protein Structure with Geometric Vector Perceptrons
[ Paper ]
Oral
Tue May 04 01:13 PM -- 01:28 PM (PDT)
On the mapping between Hopfield networks and Restricted Boltzmann Machines
[ Paper ]
Spotlight
Tue May 04 01:28 PM -- 01:38 PM (PDT)
Learning-based Support Estimation in Sublinear Time
[ Paper ]
Expo Talk Panel
Tue May 04 02:00 PM -- 03:00 PM (PDT)
Interpretability with skeptical and user-centric mind
Break
Tue May 04 03:00 PM -- 04:00 PM (PDT)
BREAK:
Town Hall
Tue May 04 04:00 PM -- 05:00 PM (PDT)
Town Hall
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Learning Safe Multi-agent Control with Decentralized Neural Barrier Certificates
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Dataset Inference: Ownership Resolution in Machine Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Achieving Linear Speedup with Partial Worker Participation in Non-IID Federated Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Gradient Descent on Neural Networks Typically Occurs at the Edge of Stability
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Fantastic Four: Differentiable and Efficient Bounds on Singular Values of Convolution Layers
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Contextual Dropout: An Efficient Sample-Dependent Dropout Module
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Learning to Reach Goals via Iterated Supervised Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Mirostat: A Neural Text Decoding Algorithm That Directly Controls Perplexity
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
CopulaGNN: Towards Integrating Representational and Correlational Roles of Graphs in Graph Neural Networks
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
CoDA: Contrast-enhanced and Diversity-promoting Data Augmentation for Natural Language Understanding
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
DDPNOpt: Differential Dynamic Programming Neural Optimizer
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Neural Mechanics: Symmetry and Broken Conservation Laws in Deep Learning Dynamics
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Debiasing Concept-based Explanations with Causal Analysis
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
A Temporal Kernel Approach for Deep Learning with Continuous-time Information
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
DOP: Off-Policy Multi-Agent Decomposed Policy Gradients
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Mathematical Reasoning via Self-supervised Skip-tree Training
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Implicit Convex Regularizers of CNN Architectures: Convex Optimization of Two- and Three-Layer Networks in Polynomial Time
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
DrNAS: Dirichlet Neural Architecture Search
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Unsupervised Discovery of 3D Physical Objects
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Co-Mixup: Saliency Guided Joint Mixup with Supermodular Diversity
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Fourier Neural Operator for Parametric Partial Differential Equations
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
The Importance of Pessimism in Fixed-Dataset Policy Optimization
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Hopper: Multi-hop Transformer for Spatiotemporal Reasoning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Hierarchical Autoregressive Modeling for Neural Video Compression
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Drop-Bottleneck: Learning Discrete Compressed Representation for Noise-Robust Exploration
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Diverse Video Generation using a Gaussian Process Trigger
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Attentional Constellation Nets for Few-Shot Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
A Discriminative Gaussian Mixture Model with Sparsity
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
HalentNet: Multimodal Trajectory Forecasting with Hallucinative Intents
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Viewmaker Networks: Learning Views for Unsupervised Representation Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Bowtie Networks: Generative Modeling for Joint Few-Shot Recognition and Novel-View Synthesis
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Fuzzy Tiling Activations: A Simple Approach to Learning Sparse Representations Online
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
RNNLogic: Learning Logic Rules for Reasoning on Knowledge Graphs
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Can a Fruit Fly Learn Word Embeddings?
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
How Neural Networks Extrapolate: From Feedforward to Graph Neural Networks
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Understanding the role of importance weighting for deep learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
FedBE: Making Bayesian Model Ensemble Applicable to Federated Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Large Batch Simulation for Deep Reinforcement Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
SEDONA: Search for Decoupled Neural Networks toward Greedy Block-wise Learning
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Lipschitz Recurrent Neural Networks
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Generating Adversarial Computer Programs using Optimized Obfuscations
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Are Neural Rankers still Outperformed by Gradient Boosted Decision Trees?
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Topology-Aware Segmentation Using Discrete Morse Theory
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
CompOFA – Compound Once-For-All Networks for Faster Multi-Platform Deployment
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Multi-resolution modeling of a discrete stochastic process identifies causes of cancer
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
A Hypergradient Approach to Robust Regression without Correspondence
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
BUSTLE: Bottom-Up Program Synthesis Through Learning-Guided Exploration
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Do Wide and Deep Networks Learn the Same Things? Uncovering How Neural Network Representations Vary with Width and Depth
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Federated Semi-Supervised Learning with Inter-Client Consistency & Disjoint Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Convex Potential Flows: Universal Probability Distributions with Optimal Transport and Convex Optimization
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
A unifying view on implicit bias in training linear neural networks
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
On Graph Neural Networks versus Graph-Augmented MLPs
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Autoregressive Dynamics Models for Offline Policy Evaluation and Optimization
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Enjoy Your Editing: Controllable GANs for Image Editing via Latent Space Navigation
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Why Are Convolutional Nets More Sample-Efficient than Fully-Connected Nets?
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
RMSprop converges with proper hyper-parameter
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Discovering Non-monotonic Autoregressive Orderings with Variational Inference
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Implicit Under-Parameterization Inhibits Data-Efficient Deep Reinforcement Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Usable Information and Evolution of Optimal Representations During Training
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
How to Find Your Friendly Neighborhood: Graph Attention Design with Self-Supervision
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Linear Mode Connectivity in Multitask and Continual Learning
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
CcGAN: Continuous Conditional Generative Adversarial Networks for Image Generation
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Knowledge Distillation as Semiparametric Inference
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Deep Equals Shallow for ReLU Networks in Kernel Regimes
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Projected Latent Markov Chain Monte Carlo: Conditional Sampling of Normalizing Flows
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT)
Discrete Graph Structure Learning for Forecasting Multiple Time Series
[ Paper ]
Poster
Tue May 04 05:00 PM -- 07:00 PM (PDT) @ Virtual
Universal Weakly Supervised Segmentation by Pixel-to-Segment Contrastive Learning
Oral
Tue May 04 07:00 PM -- 07:15 PM (PDT)
Deep symbolic regression: Recovering mathematical expressions from data via risk-seeking policy gradients
[ Paper ]
Spotlight
Tue May 04 07:15 PM -- 07:25 PM (PDT)
DDPNOpt: Differential Dynamic Programming Neural Optimizer
[ Paper ]
Spotlight
Tue May 04 07:25 PM -- 07:35 PM (PDT)
Orthogonalizing Convolutional Layers with the Cayley Transform
[ Paper ]
Spotlight
Tue May 04 07:35 PM -- 07:45 PM (PDT)
Model-Based Visual Planning with Self-Supervised Functional Distances
[ Paper ]
Oral
Tue May 04 07:55 PM -- 08:10 PM (PDT)
Global Convergence of Three-layer Neural Networks in the Mean Field Regime
[ Paper ]
Spotlight
Tue May 04 08:20 PM -- 08:30 PM (PDT)
Async-RED: A Provably Convergent Asynchronous Block Parallel Stochastic Method using Deep Denoising Priors
[ Paper ]
Spotlight
Tue May 04 08:40 PM -- 08:50 PM (PDT)
Are Neural Rankers still Outperformed by Gradient Boosted Decision Trees?
[ Paper ]
Oral
Tue May 04 09:03 PM -- 09:18 PM (PDT)
Co-Mixup: Saliency Guided Joint Mixup with Supermodular Diversity
[ Paper ]
Oral
Tue May 04 09:18 PM -- 09:33 PM (PDT)
MONGOOSE: A Learnable LSH Framework for Efficient Neural Network Training
[ Paper ]
Spotlight
Tue May 04 09:33 PM -- 09:43 PM (PDT)
Locally Free Weight Sharing for Network Width Search
[ Paper ]
Break
Tue May 04 10:30 PM -- 10:30 PM (PDT)
BREAK:
Invited Talk
Wed May 05 12:00 AM -- 01:00 AM (PDT) @ Virtual
Perceiving the 3D World from Images and Video
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Differentiable Segmentation of Sequences
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Removing Undesirable Feature Contributions Using Out-of-Distribution Data
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Net-DNF: Effective Deep Modeling of Tabular Data
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstruction
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
A Wigner-Eckart Theorem for Group Equivariant Convolution Kernels
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Negative Data Augmentation
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Expressive Power of Invariant and Equivariant Graph Neural Networks
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Long Range Arena : A Benchmark for Efficient Transformers
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
DICE: Diversity in Deep Ensembles via Conditional Redundancy Adversarial Estimation
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Efficient Continual Learning with Modular Networks and Task-Driven Priors
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Reweighting Augmented Samples by Minimizing the Maximal Expected Loss
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Relating by Contrasting: A Data-efficient Framework for Multimodal Generative Models
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Self-supervised Adversarial Robustness for the Low-label, High-data Regime
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Return-Based Contrastive Representation Learning for Reinforcement Learning
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Learning from Demonstration with Weakly Supervised Disentanglement
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
On Data-Augmentation and Consistency-Based Semi-Supervised Learning
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Degree-Quant: Quantization-Aware Training for Graph Neural Networks
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Learning Associative Inference Using Fast Weight Memory
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Improving Relational Regularized Autoencoders with Spherical Sliced Fused Gromov Wasserstein
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Identifying Physical Law of Hamiltonian Systems via Meta-Learning
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Discovering Diverse Multi-Agent Strategic Behavior via Reward Randomization
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Byzantine-Resilient Non-Convex Stochastic Gradient Descent
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Neural ODE Processes
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Robust Learning of Fixed-Structure Bayesian Networks in Nearly-Linear Time
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Learning Task Decomposition with Ordered Memory Policy Network
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
FOCAL: Efficient Fully-Offline Meta-Reinforcement Learning via Distance Metric Learning and Behavior Regularization
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Disentangled Recurrent Wasserstein Autoencoder
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Active Contrastive Learning of Audio-Visual Video Representations
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Communication in Multi-Agent Reinforcement Learning: Intention Sharing
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Bidirectional Variational Inference for Non-Autoregressive Text-to-Speech
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Deployment-Efficient Reinforcement Learning via Model-Based Offline Optimization
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Interpretable Neural Architecture Search via Bayesian Optimisation with Weisfeiler-Lehman Kernels
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Acting in Delayed Environments with Non-Stationary Markov Policies
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Deep Neural Network Fingerprinting by Conferrable Adversarial Examples
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Separation and Concentration in Deep Networks
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Locally Free Weight Sharing for Network Width Search
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Improving Adversarial Robustness via Channel-wise Activation Suppressing
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Explaining by Imitating: Understanding Decisions by Interpretable Policy Learning
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
A Better Alternative to Error Feedback for Communication-Efficient Distributed Learning
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
IEPT: Instance-Level and Episode-Level Pretext Tasks for Few-Shot Learning
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Dance Revolution: Long-Term Dance Generation with Music via Curriculum Learning
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Neural Delay Differential Equations
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Geometry-aware Instance-reweighted Adversarial Training
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Bag of Tricks for Adversarial Training
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
CoCon: A Self-Supervised Approach for Controlled Text Generation
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
No Cost Likelihood Manipulation at Test Time for Making Better Mistakes in Deep Networks
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
New Bounds For Distributed Mean Estimation and Variance Reduction
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Neural gradients are near-lognormal: improved quantized and sparse training
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Isometric Propagation Network for Generalized Zero-shot Learning
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT) @ Virtual
Graph Edit Networks
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Auxiliary Task Update Decomposition: The Good, the Bad and the Neutral
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting
[ Paper ]
Poster
Wed May 05 01:00 AM -- 03:00 AM (PDT)
Knowledge distillation via softmax regression representation learning
[ Paper ]
Oral
Wed May 05 03:00 AM -- 03:15 AM (PDT)
An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
[ Paper ]
Oral
Wed May 05 03:30 AM -- 03:45 AM (PDT)
Share or Not? Learning to Schedule Language-Specific Capacity for Multilingual Translation
[ Paper ]
Spotlight
Wed May 05 03:45 AM -- 03:55 AM (PDT)
Support-set bottlenecks for video-text representation learning
[ Paper ]
Oral
Wed May 05 04:05 AM -- 04:20 AM (PDT)
Getting a CLUE: A Method for Explaining Uncertainty Estimates
[ Paper ]
Spotlight
Wed May 05 04:20 AM -- 04:30 AM (PDT)
Influence Estimation for Generative Adversarial Networks
[ Paper ]
Spotlight
Wed May 05 04:40 AM -- 04:50 AM (PDT)
Deep Neural Network Fingerprinting by Conferrable Adversarial Examples
[ Paper ]
Oral
Wed May 05 05:00 AM -- 05:15 AM (PDT)
Learning Cross-Domain Correspondence for Control with Dynamics Cycle-Consistency
[ Paper ]
Spotlight
Wed May 05 05:15 AM -- 05:25 AM (PDT)
Benefit of deep learning with non-convex noisy gradient descent: Provable excess risk bound and superiority to kernel methods
[ Paper ]
Spotlight
Wed May 05 05:25 AM -- 05:35 AM (PDT)
Tent: Fully Test-Time Adaptation by Entropy Minimization
[ Paper ]
Spotlight
Wed May 05 05:35 AM -- 05:45 AM (PDT)
Neural Approximate Sufficient Statistics for Implicit Models
[ Paper ]
Break
Wed May 05 06:15 AM -- 08:00 AM (PDT)
BREAK:
Invited Talk
Wed May 05 08:00 AM -- 09:00 AM (PDT) @ Virtual
Is My Dataset Biased?
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Learning advanced mathematical computations from examples
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Neural Synthesis of Binaural Speech From Mono Audio
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Early Stopping in Deep Networks: Double Descent and How to Eliminate it
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Learning Generalizable Visual Representations via Interactive Gameplay
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
DeepAveragers: Offline Reinforcement Learning By Solving Derived Non-Parametric MDPs
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Neural Networks for Learning Counterfactual G-Invariances from Single Environments
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Ask Your Humans: Using Human Instructions to Improve Generalization in Reinforcement Learning
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Long-tail learning via logit adjustment
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
You Only Need Adversarial Supervision for Semantic Image Synthesis
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Probabilistic Numeric Convolutional Neural Networks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
How Does Mixup Help With Robustness and Generalization?
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
For self-supervised learning, Rationality implies generalization, provably
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Coupled Oscillatory Recurrent Neural Network (coRNN): An accurate and (gradient) stable architecture for learning long time dependencies
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Learning to Represent Action Values as a Hypergraph on the Action Vertices
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Multivariate Probabilistic Time Series Forecasting via Conditioned Normalizing Flows
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Self-supervised Visual Reinforcement Learning with Object-centric Representations
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Optimism in Reinforcement Learning with Generalized Linear Function Approximation
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Entropic gradient descent algorithms and wide flat minima
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
RODE: Learning Roles to Decompose Multi-Agent Tasks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Multiplicative Filter Networks
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Property Controllable Variational Autoencoder via Invertible Mutual Dependence
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
TropEx: An Algorithm for Extracting Linear Terms in Deep Neural Networks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Continuous Wasserstein-2 Barycenter Estimation without Minimax Optimization
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Growing Efficient Deep Networks by Structured Continuous Sparsification
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
SEED: Self-supervised Distillation For Visual Representation
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Unsupervised Audiovisual Synthesis via Exemplar Autoencoders
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
On the Bottleneck of Graph Neural Networks and its Practical Implications
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Theoretical bounds on estimation error for meta-learning
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Deep symbolic regression: Recovering mathematical expressions from data via risk-seeking policy gradients
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Graph Traversal with Tensor Functionals: A Meta-Algorithm for Scalable Learning
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Learning Task-General Representations with Generative Neuro-Symbolic Modeling
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Evaluation of Neural Architectures Trained With Square Loss vs Cross-Entropy in Classification Tasks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Chaos of Learning Beyond Zero-sum and Coordination via Game Decompositions
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
A Gradient Flow Framework For Analyzing Network Pruning
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Sharpness-aware Minimization for Efficiently Improving Generalization
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
My Body is a Cage: the Role of Morphology in Graph-Based Incompatible Control
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Interpreting Graph Neural Networks for NLP With Differentiable Edge Masking
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Modeling the Second Player in Distributionally Robust Optimization
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Are Neural Nets Modular? Inspecting Functional Modularity Through Differentiable Weight Masks
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Variational State-Space Models for Localisation and Dense 3D Mapping in 6 DoF
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Few-Shot Bayesian Optimization with Deep Kernel Surrogates
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Faster Binary Embeddings for Preserving Euclidean Distances
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Provably robust classification of adversarial examples with detection
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
NAS-Bench-ASR: Reproducible Neural Architecture Search for Speech Recognition
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Human-Level Performance in No-Press Diplomacy via Equilibrium Search
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Witches' Brew: Industrial Scale Data Poisoning via Gradient Matching
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
OPAL: Offline Primitive Discovery for Accelerating Offline Reinforcement Learning
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Geometry-Aware Gradient Algorithms for Neural Architecture Search
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
HyperDynamics: Meta-Learning Object and Agent Dynamics with Hypernetworks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Average-case Acceleration for Bilinear Games and Normal Matrices
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Pre-training Text-to-Text Transformers for Concept-centric Common Sense
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Generative Time-series Modeling with Fourier Flows
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Learning Mesh-Based Simulation with Graph Networks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Anchor & Transform: Learning Sparse Embeddings for Large Vocabularies
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Training independent subnetworks for robust prediction
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Federated Learning via Posterior Averaging: A New Perspective and Practical Algorithms
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Graph Information Bottleneck for Subgraph Recognition
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Exploring the Uncertainty Properties of Neural Networks’ Implicit Priors in the Infinite-Width Limit
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
IsarStep: a Benchmark for High-level Mathematical Reasoning
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
HyperGrid Transformers: Towards A Single Model for Multiple Tasks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Iterated learning for emergent systematicity in VQA
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Differentiable Trust Region Layers for Deep Reinforcement Learning
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Orthogonalizing Convolutional Layers with the Cayley Transform
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
PC2WF: 3D Wireframe Reconstruction from Raw Point Clouds
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Perceptual Adversarial Robustness: Defense Against Unseen Threat Models
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
More or Less: When and How to Build Convolutional Neural Network Ensembles
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
DARTS-: Robustly Stepping out of Performance Collapse Without Indicators
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Watch-And-Help: A Challenge for Social Perception and Human-AI Collaboration
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Boost then Convolve: Gradient Boosting Meets Graph Neural Networks
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT)
Unbiased Teacher for Semi-Supervised Object Detection
[ Paper ]
Poster
Wed May 05 09:00 AM -- 11:00 AM (PDT) @ Virtual
Anytime Sampling for Autoregressive Models via Ordered Autoencoding
Oral
Wed May 05 11:00 AM -- 11:15 AM (PDT)
Human-Level Performance in No-Press Diplomacy via Equilibrium Search
[ Paper ]
Oral
Wed May 05 11:15 AM -- 11:30 AM (PDT)
Learning to Reach Goals via Iterated Supervised Learning
[ Paper ]
Oral
Wed May 05 11:30 AM -- 11:45 AM (PDT)
Learning Invariant Representations for Reinforcement Learning without Reconstruction
[ Paper ]
Spotlight
Wed May 05 12:00 PM -- 12:10 PM (PDT)
Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
[ Paper ]
Oral
Wed May 05 12:23 PM -- 12:38 PM (PDT)
Coupled Oscillatory Recurrent Neural Network (coRNN): An accurate and (gradient) stable architecture for learning long time dependencies
[ Paper ]
Spotlight
Wed May 05 12:38 PM -- 12:48 PM (PDT)
Sequential Density Ratio Estimation for Simultaneous Optimization of Speed and Accuracy
[ Paper ]
Spotlight
Wed May 05 12:48 PM -- 12:58 PM (PDT)
LambdaNetworks: Modeling long-range Interactions without Attention
[ Paper ]
Spotlight
Wed May 05 01:18 PM -- 01:28 PM (PDT)
Unsupervised Object Keypoint Learning using Local Spatial Predictability
[ Paper ]
Spotlight
Wed May 05 01:28 PM -- 01:38 PM (PDT)
VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models
[ Paper ]
Spotlight
Wed May 05 01:48 PM -- 01:58 PM (PDT)
A Gradient Flow Framework For Analyzing Network Pruning
[ Paper ]
Spotlight
Wed May 05 01:58 PM -- 02:08 PM (PDT)
Differentially Private Learning Needs Better Features (or Much More Data)
[ Paper ]
Expo Talk Panel
Wed May 05 02:00 PM -- 03:00 PM (PDT)
Live Panel - Academics@ Presents: Representation Learning at Amazon
Break
Wed May 05 03:00 PM -- 04:00 PM (PDT)
BREAK:
Oral
Wed May 05 04:00 PM -- 04:15 PM (PDT)
Neural Synthesis of Binaural Speech From Mono Audio
[ Paper ]
Oral
Wed May 05 04:30 PM -- 04:45 PM (PDT)
Score-Based Generative Modeling through Stochastic Differential Equations
[ Paper ]
Spotlight
Wed May 05 04:45 PM -- 04:55 PM (PDT)
Learning Mesh-Based Simulation with Graph Networks
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Efficient Conformal Prediction via Cascaded Inference with Expanded Admission
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
gradSim: Differentiable simulation for system identification and visuomotor control
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Evaluating the Disentanglement of Deep Generative Models through Manifold Topology
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Influence Functions in Deep Learning Are Fragile
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Filtered Inner Product Projection for Crosslingual Embedding Alignment
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
A Geometric Analysis of Deep Generative Image Models and Its Applications
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Learning Manifold Patch-Based Representations of Man-Made Shapes
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Measuring Massive Multitask Language Understanding
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Estimating informativeness of samples with Smooth Unique Information
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
DiffWave: A Versatile Diffusion Model for Audio Synthesis
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Generating Furry Cars: Disentangling Object Shape and Appearance across Multiple Domains
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
PolarNet: Learning to Optimize Polar Keypoints for Keypoint Based Object Detection
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Learning to Generate 3D Shapes with Generative Cellular Automata
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
BERTology Meets Biology: Interpreting Attention in Protein Language Models
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Explainable Subgraph Reasoning for Forecasting on Temporal Knowledge Graphs
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Revisiting Dynamic Convolution via Matrix Decomposition
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Learning to Deceive Knowledge Graph Augmented Models via Targeted Perturbation
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Neural Architecture Search on ImageNet in Four GPU Hours: A Theoretically Inspired Perspective
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Adaptive Procedural Task Generation for Hard-Exploration Problems
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Learning and Evaluating Representations for Deep One-Class Classification
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Estimating Lipschitz constants of monotone deep equilibrium models
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
A PAC-Bayesian Approach to Generalization Bounds for Graph Neural Networks
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Mind the Pad -- CNNs Can Develop Blind Spots
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Emergent Symbols through Binding in External Memory
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Interactive Weak Supervision: Learning Useful Heuristics for Data Labeling
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
AdaSpeech: Adaptive Text to Speech for Custom Voice
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Simple Augmentation Goes a Long Way: ADRL for DNN Quantization
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Efficient Empowerment Estimation for Unsupervised Stabilization
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Evolving Reinforcement Learning Algorithms
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Deep Encoder, Shallow Decoder: Reevaluating Non-autoregressive Machine Translation
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
AutoLRS: Automatic Learning-Rate Schedule by Bayesian Optimization on the Fly
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Robust Overfitting may be mitigated by properly learned smoothening
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Learning Long-term Visual Dynamics with Region Proposal Interaction Networks
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Improved Autoregressive Modeling with Distribution Smoothing
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Beyond Categorical Label Representations for Image Classification
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Efficient Reinforcement Learning in Factored MDPs with Application to Constrained RL
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Learning with Instance-Dependent Label Noise: A Sample Sieve Approach
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Local Convergence Analysis of Gradient Descent Ascent with Finite Timescale Separation
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
CoCo: Controllable Counterfactuals for Evaluating Dialogue State Trackers
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Wandering within a world: Online contextualized few-shot learning
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Modelling Hierarchical Structure between Dialogue Policy and Natural Language Generator with Option Framework for Task-oriented Dialogue System
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Graph-Based Continual Learning
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Effective and Efficient Vote Attack on Capsule Networks
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Economic Hyperparameter Optimization With Blended Search Strategy
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Learning with Feature-Dependent Label Noise: A Progressive Approach
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
On the Critical Role of Conventions in Adaptive Human-AI Collaboration
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Is Attention Better Than Matrix Decomposition?
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Protecting DNNs from Theft using an Ensemble of Diverse Models
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
INT: An Inequality Benchmark for Evaluating Generalization in Theorem Proving
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Deep Neural Tangent Kernel and Laplace Kernel Have the Same RKHS
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Empirical or Invariant Risk Minimization? A Sample Complexity Perspective
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Control-Aware Representations for Model-based Reinforcement Learning
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Mixed-Features Vectors and Subspace Splitting
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Fast And Slow Learning Of Recurrent Independent Mechanisms
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Evaluations and Methods for Explanation through Robustness Analysis
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Combining Physics and Machine Learning for Network Flow Estimation
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Efficient Wasserstein Natural Gradients for Reinforcement Learning
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
Adaptive Universal Generalized PageRank Graph Neural Network
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Better Fine-Tuning by Reducing Representational Collapse
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Empirical Analysis of Unlabeled Entity Problem in Named Entity Recognition
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
Model-Based Visual Planning with Self-Supervised Functional Distances
[ Paper ]
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
CPR: Classifier-Projection Regularization for Continual Learning
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT) @ Virtual
In-N-Out: Pre-Training and Self-Training using Auxiliary Information for Out-of-Distribution Robustness
Poster
Wed May 05 05:00 PM -- 07:00 PM (PDT)
AdaFuse: Adaptive Temporal Fusion Network for Efficient Action Recognition
[ Paper ]
Oral
Wed May 05 07:00 PM -- 07:15 PM (PDT)
Improved Autoregressive Modeling with Distribution Smoothing
[ Paper ]
Spotlight
Wed May 05 07:25 PM -- 07:35 PM (PDT)
Large Scale Image Completion via Co-Modulated Generative Adversarial Networks
[ Paper ]
Spotlight
Wed May 05 07:35 PM -- 07:45 PM (PDT)
Emergent Symbols through Binding in External Memory
[ Paper ]
Oral
Wed May 05 07:55 PM -- 08:10 PM (PDT)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
[ Paper ]
Spotlight
Wed May 05 08:20 PM -- 08:30 PM (PDT)
Understanding the role of importance weighting for deep learning
[ Paper ]
Spotlight
Wed May 05 08:30 PM -- 08:40 PM (PDT)
Towards Robustness Against Natural Language Word Substitutions
[ Paper ]
Spotlight
Wed May 05 08:40 PM -- 08:50 PM (PDT)
Undistillable: Making A Nasty Teacher That CANNOT teach students
[ Paper ]
Spotlight
Wed May 05 08:50 PM -- 09:00 PM (PDT)
CPT: Efficient Deep Neural Network Training via Cyclic Precision
[ Paper ]
Spotlight
Wed May 05 09:15 PM -- 09:25 PM (PDT)
PlasticineLab: A Soft-Body Manipulation Benchmark with Differentiable Physics
[ Paper ]
Spotlight
Wed May 05 09:25 PM -- 09:35 PM (PDT)
Regularization Matters in Policy Optimization - An Empirical Study on Continuous Control
[ Paper ]
Spotlight
Wed May 05 09:35 PM -- 09:45 PM (PDT)
Regularized Inverse Reinforcement Learning
[ Paper ]
Spotlight
Wed May 05 09:45 PM -- 09:55 PM (PDT)
Behavioral Cloning from Noisy Demonstrations
[ Paper ]
Break
Wed May 05 10:15 PM -- 10:15 PM (PDT)
BREAK:
Oral
Thu May 06 12:00 AM -- 12:15 AM (PDT)
Rethinking Architecture Selection in Differentiable NAS
[ Paper ]
Session
Thu May 06 12:00 AM -- 01:05 AM (PDT)
Outstanding Paper Session 2
Oral
Thu May 06 12:15 AM -- 12:30 AM (PDT)
Complex Query Answering with Neural Link Predictors
[ Paper ]
Oral
Thu May 06 12:30 AM -- 12:45 AM (PDT)
Optimal Rates for Averaged Stochastic Gradient Descent under Neural Tangent Kernel Regime
[ Paper ]
Spotlight
Thu May 06 12:45 AM -- 12:55 AM (PDT)
Beyond Fully-Connected Layers with Quaternions: Parameterization of Hypercomplex Multiplications with $1/n$ Parameters
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Understanding the effects of data parallelism and sparsity on neural network training
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Genetic Soft Updates for Policy Evolution in Deep Reinforcement Learning
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Learning continuous-time PDEs from sparse data with graph neural networks
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
What Matters for On-Policy Deep Actor-Critic Methods? A Large-Scale Study
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Influence Estimation for Generative Adversarial Networks
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Learning Deep Features in Instrumental Variable Regression
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Network Pruning That Matters: A Case Study on Retraining Variants
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Kanerva++: Extending the Kanerva Machine With Differentiable, Locally Block Allocated Latent Memory
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Incremental few-shot learning via vector quantization in deep embedded space
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
An Unsupervised Deep Learning Approach for Real-World Image Denoising
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Contrastive Divergence Learning is a Time Reversal Adversarial Game
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
IOT: Instance-wise Layer Reordering for Transformer Structures
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
The inductive bias of ReLU networks on orthogonally separable data
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Certify or Predict: Boosting Certified Robustness with Compositional Architectures
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Private Image Reconstruction from System Side Channels Using Generative Models
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
AdamP: Slowing Down the Slowdown for Momentum Optimizers on Scale-invariant Weights
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Learnable Embedding sizes for Recommender Systems
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Colorization Transformer
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Do not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Loss Function Discovery for Object Detection via Convergence-Simulation Driven Search
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Distilling Knowledge from Reader to Retriever for Question Answering
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Adaptive Extra-Gradient Methods for Min-Max Optimization and Games
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
ChipNet: Budget-Aware Pruning with Heaviside Continuous Approximations
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Conditional Generative Modeling via Learning the Latent Space
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Contrastive Learning with Adversarial Perturbations for Conditional Text Generation
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Collective Robustness Certificates: Exploiting Interdependence in Graph Neural Networks
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Emergent Road Rules In Multi-Agent Driving Environments
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Scalable Learning and MAP Inference for Nonsymmetric Determinantal Point Processes
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Representation Balancing Offline Model-based Reinforcement Learning
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Inductive Representation Learning in Temporal Networks via Causal Anonymous Walks
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Efficient Inference of Flexible Interaction in Spiking-neuron Networks
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Optimal Conversion of Conventional Artificial Neural Networks to Spiking Neural Networks
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Balancing Constraints and Rewards with Meta-Gradient D4PG
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Free Lunch for Few-shot Learning: Distribution Calibration
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Learning Neural Generative Dynamics for Molecular Conformation Generation
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
A Diffusion Theory For Deep Learning Dynamics: Stochastic Gradient Descent Exponentially Favors Flat Minima
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Categorical Normalizing Flows via Continuous Transformations
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Mind the Gap when Conditioning Amortised Inference in Sequential Latent-Variable Models
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Optimal Rates for Averaged Stochastic Gradient Descent under Neural Tangent Kernel Regime
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Impact of Representation Learning in Linear Bandits
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
CausalWorld: A Robotic Manipulation Benchmark for Causal Structure and Transfer Learning
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Practical Massively Parallel Monte-Carlo Tree Search Applied to Molecular Design
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Sparse Quantized Spectral Clustering
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Continual learning in recurrent neural networks
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Revisiting Hierarchical Approach for Persistent Long-Term Video Prediction
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Interpretable Models for Granger Causality Using Self-explaining Neural Networks
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Retrieval-Augmented Generation for Code Summarization via Hybrid GNN
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Grounding Language to Autonomously-Acquired Skills via Goal Generation
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Go with the flow: Adaptive control for Neural ODEs
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT) @ Virtual
Learning Reasoning Paths over Semantic Graphs for Video-grounded Dialogues
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Deformable DETR: Deformable Transformers for End-to-End Object Detection
[ Paper ]
Poster
Thu May 06 01:00 AM -- 03:00 AM (PDT)
Contrastive Explanations for Reinforcement Learning via Embedded Self Predictions
[ Paper ]
Oral
Thu May 06 03:00 AM -- 03:15 AM (PDT)
What Matters for On-Policy Deep Actor-Critic Methods? A Large-Scale Study
[ Paper ]
Spotlight
Thu May 06 03:15 AM -- 03:25 AM (PDT)
Winning the L2RPN Challenge: Power Grid Management via Semi-Markov Afterstate Actor-Critic
[ Paper ]
Spotlight
Thu May 06 03:25 AM -- 03:35 AM (PDT)
UPDeT: Universal Multi-agent RL via Policy Decoupling with Transformers
[ Paper ]
Spotlight
Thu May 06 03:35 AM -- 03:45 AM (PDT)
Quantifying Differences in Reward Functions
[ Paper ]
Spotlight
Thu May 06 03:45 AM -- 03:55 AM (PDT)
Iterative Empirical Game Solving via Single Policy Best Response
[ Paper ]
Spotlight
Thu May 06 03:55 AM -- 04:05 AM (PDT)
Discovering a set of policies for the worst case reward
[ Paper ]
Oral
Thu May 06 04:20 AM -- 04:35 AM (PDT)
Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting
[ Paper ]
Spotlight
Thu May 06 04:35 AM -- 04:45 AM (PDT)
Unlearnable Examples: Making Personal Data Unexploitable
[ Paper ]
Spotlight
Thu May 06 04:45 AM -- 04:55 AM (PDT)
Self-supervised Visual Reinforcement Learning with Object-centric Representations
[ Paper ]
Spotlight
Thu May 06 04:55 AM -- 05:05 AM (PDT)
On Self-Supervised Image Representations for GAN Evaluation
[ Paper ]
Spotlight
Thu May 06 05:05 AM -- 05:15 AM (PDT)
Retrieval-Augmented Generation for Code Summarization via Hybrid GNN
[ Paper ]
Spotlight
Thu May 06 05:15 AM -- 05:25 AM (PDT)
Practical Real Time Recurrent Learning with a Sparse Approximation
[ Paper ]
Break
Thu May 06 06:00 AM -- 08:00 AM (PDT)
BREAK:
Invited Talk
Thu May 06 08:00 AM -- 09:00 AM (PDT) @ Virtual
Soft bodied robots for human centered design of robots for everyday life
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Understanding and Improving Encoder Layer Fusion in Sequence-to-Sequence Learning
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Linear Last-iterate Convergence in Constrained Saddle-point Optimization
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Integrating Categorical Semantics into Unsupervised Domain Translation
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Understanding and Improving Lexical Choice in Non-Autoregressive Translation
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
C-Learning: Learning to Achieve Goals via Recursive Classification
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Adversarial score matching and improved sampling for image generation
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Differentially Private Learning Needs Better Features (or Much More Data)
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Learning to live with Dale's principle: ANNs with separate excitatory and inhibitory units
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
A teacher-student framework to distill future trajectories
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
On Position Embeddings in BERT
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Correcting experience replay for multi-agent communication
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Adversarially-Trained Deep Nets Transfer Better: Illustration on Image Classification
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Improving Zero-Shot Voice Style Transfer via Disentangled Representation Learning
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Graph Coarsening with Neural Networks
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Neural Learning of One-of-Many Solutions for Combinatorial Problems in Structured Output Spaces
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Uncertainty in Gradient Boosting via Ensembles
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Multi-Class Uncertainty Calibration via Mutual Information Maximization-based Binning
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Generative Language-Grounded Policy in Vision-and-Language Navigation with Bayes' Rule
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Cut out the annotator, keep the cutout: better segmentation with weak supervision
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Learning to Recombine and Resample Data For Compositional Generalization
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
A Critique of Self-Expressive Deep Subspace Clustering
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Efficient Transformers in Reinforcement Learning using Actor-Learner Distillation
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Deep Networks and the Multiple Manifold Problem
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Initialization and Regularization of Factorized Neural Layers
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Enforcing robust control guarantees within neural network policies
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Domain-Robust Visual Imitation Learning with Mutual Information Constraints
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Blending MPC & Value Function Approximation for Efficient Reinforcement Learning
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Deconstructing the Regularization of BatchNorm
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Noise or Signal: The Role of Image Backgrounds in Object Recognition
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Learning to Set Waypoints for Audio-Visual Navigation
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Lifelong Learning of Compositional Structures
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
VCNet and Functional Targeted Regularization For Learning Causal Effects of Continuous Treatments
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Learning Cross-Domain Correspondence for Control with Dynamics Cycle-Consistency
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Dataset Meta-Learning from Kernel Ridge-Regression
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Rethinking Soft Labels for Knowledge Distillation: A Bias–Variance Tradeoff Perspective
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Meta-learning with negative learning rates
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Hierarchical Reinforcement Learning by Discovering Intrinsic Options
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Contrastive Learning with Hard Negative Samples
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Dynamic Tensor Rematerialization
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learning
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
EEC: Learning to Encode and Regenerate Images for Continual Learning
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Improving VAEs' Robustness to Adversarial Attack
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Gradient Vaccine: Investigating and Improving Multi-task Optimization in Massively Multilingual Models
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Variational Information Bottleneck for Effective Low-Resource Fine-Tuning
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Dual-mode ASR: Unify and Improve Streaming ASR with Full-context Modeling
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Analyzing the Expressive Power of Graph Neural Networks in a Spectral Perspective
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Meta-Learning of Structured Task Distributions in Humans and Machines
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
BSQ: Exploring Bit-Level Sparsity for Mixed-Precision Neural Network Quantization
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Robust early-learning: Hindering the memorization of noisy labels
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Local Search Algorithms for Rank-Constrained Convex Optimization
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
A Mathematical Exploration of Why Language Models Help Solve Downstream Tasks
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
CaPC Learning: Confidential and Private Collaborative Learning
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
Bayesian Few-Shot Classification with One-vs-Each Pólya-Gamma Augmented Gaussian Processes
[ Paper ]
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT) @ Virtual
Heating up decision boundaries: isocapacitory saturation, adversarial scenarios and generalization bounds
Poster
Thu May 06 09:00 AM -- 11:00 AM (PDT)
PMI-Masking: Principled masking of correlated spans
[ Paper ]
Oral
Thu May 06 11:00 AM -- 11:15 AM (PDT)
VCNet and Functional Targeted Regularization For Learning Causal Effects of Continuous Treatments
[ Paper ]
Oral
Thu May 06 11:15 AM -- 11:30 AM (PDT)
SenSeI: Sensitive Set Invariance for Enforcing Individual Fairness
[ Paper ]
Oral
Thu May 06 11:45 AM -- 12:00 PM (PDT)
Why Are Convolutional Nets More Sample-Efficient than Fully-Connected Nets?
[ Paper ]
Spotlight
Thu May 06 12:10 PM -- 12:20 PM (PDT)
Correcting experience replay for multi-agent communication
[ Paper ]
Spotlight
Thu May 06 12:20 PM -- 12:30 PM (PDT)
Contrastive Behavioral Similarity Embeddings for Generalization in Reinforcement Learning
[ Paper ]
Spotlight
Thu May 06 12:30 PM -- 12:40 PM (PDT)
DeepAveragers: Offline Reinforcement Learning By Solving Derived Non-Parametric MDPs
[ Paper ]
Spotlight
Thu May 06 12:40 PM -- 12:50 PM (PDT)
Data-Efficient Reinforcement Learning with Self-Predictive Representations
[ Paper ]
Oral
Thu May 06 01:00 PM -- 01:15 PM (PDT)
DiffWave: A Versatile Diffusion Model for Audio Synthesis
[ Paper ]
Oral
Thu May 06 01:15 PM -- 01:30 PM (PDT)
Self-training For Few-shot Transfer Across Extreme Task Differences
[ Paper ]
Spotlight
Thu May 06 01:30 PM -- 01:40 PM (PDT)
A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference
[ Paper ]
Spotlight
Thu May 06 01:40 PM -- 01:50 PM (PDT)
BUSTLE: Bottom-Up Program Synthesis Through Learning-Guided Exploration
[ Paper ]
Spotlight
Thu May 06 01:50 PM -- 02:00 PM (PDT)
Disentangled Recurrent Wasserstein Autoencoder
[ Paper ]
Expo Talk Panel
Thu May 06 02:00 PM -- 03:00 PM (PDT)
AI Model Efficiency Toolkit talk & demo
Break
Thu May 06 03:00 PM -- 04:00 PM (PDT)
BREAK:
Invited Talk
Thu May 06 04:00 PM -- 05:00 PM (PDT) @ Virtual
Self-Supervision for Learning from the Bottom Up
[
Slides]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Few-Shot Learning via Learning the Representation, Provably
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Self-supervised Representation Learning with Relative Predictive Coding
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
LowKey: Leveraging Adversarial Attacks to Protect Social Media Users from Facial Recognition
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Contrastive Syn-to-Real Generalization
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Greedy-GQ with Variance Reduction: Finite-time Analysis and Improved Complexity
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
CPT: Efficient Deep Neural Network Training via Cyclic Precision
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Large Scale Image Completion via Co-Modulated Generative Adversarial Networks
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Answering Complex Open-Domain Questions with Multi-Hop Dense Retrieval
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
DynaTune: Dynamic Tensor Program Optimization in Deep Neural Network Compilation
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Fast and Complete: Enabling Complete Neural Network Verification with Rapid and Massively Parallel Incomplete Verifiers
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Clustering-friendly Representation Learning via Instance Discrimination and Feature Decorrelation
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
FedBN: Federated Learning on Non-IID Features via Local Batch Normalization
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
A Learning Theoretic Perspective on Local Explainability
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Generative Scene Graph Networks
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
ARMOURED: Adversarially Robust MOdels using Unlabeled data by REgularizing Diversity
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Non-asymptotic Confidence Intervals of Off-policy Evaluation: Primal and Dual Bounds
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
GAN2GAN: Generative Noise Learning for Blind Denoising with Single Noisy Images
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Molecule Optimization by Explainable Evolution
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
CO2: Consistent Contrast for Unsupervised Visual Representation Learning
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Async-RED: A Provably Convergent Asynchronous Block Parallel Stochastic Method using Deep Denoising Priors
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
MARS: Markov Molecular Sampling for Multi-objective Drug Discovery
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
On the Curse of Memory in Recurrent Neural Networks: Approximation and Optimization Analysis
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Cross-Attentional Audio-Visual Fusion for Weakly-Supervised Action Localization
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Heteroskedastic and Imbalanced Deep Learning with Adaptive Regularization
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Learning Energy-Based Generative Models via Coarse-to-Fine Expanding and Sampling
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Prototypical Representation Learning for Relation Extraction
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
$i$-Mix: A Domain-Agnostic Strategy for Contrastive Representation Learning
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Factorizing Declarative and Procedural Knowledge in Structured, Dynamical Environments
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Fully Unsupervised Diversity Denoising with Convolutional Variational Autoencoders
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Fast Geometric Projections for Local Robustness Certification
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label Selection Framework for Semi-Supervised Learning
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Adapting to Reward Progressivity via Spectral Reinforcement Learning
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Representing Partial Programs with Blended Abstract Semantics
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
FastSpeech 2: Fast and High-Quality End-to-End Text to Speech
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
BiPointNet: Binary Neural Network for Point Clouds
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Auto Seg-Loss: Searching Metric Surrogates for Semantic Segmentation
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Learning to Make Decisions via Submodular Regularization
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Group Equivariant Generative Adversarial Networks
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
CT-Net: Channel Tensorization Network for Video Classification
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
HeteroFL: Computation and Communication Efficient Federated Learning for Heterogeneous Clients
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Long-tailed Recognition by Routing Diverse Distribution-Aware Experts
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
ANOCE: Analysis of Causal Effects with Multiple Mediators via Constrained Structural Learning
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Extreme Memorization via Scale of Initialization
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Multi-timescale Representation Learning in LSTM Language Models
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Global Convergence of Three-layer Neural Networks in the Mean Field Regime
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
HW-NAS-Bench: Hardware-Aware Neural Architecture Search Benchmark
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Linear Convergent Decentralized Optimization with Compression
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Combining Ensembles and Data Augmentation Can Harm Your Calibration
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Learning to Sample with Local and Global Contexts in Experience Replay Buffer
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Combining Label Propagation and Simple Models out-performs Graph Neural Networks
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Multi-Prize Lottery Ticket Hypothesis: Finding Accurate Binary Neural Networks by Pruning A Randomly Weighted Network
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Convex Regularization behind Neural Reconstruction
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Neural representation and generation for RNA secondary structures
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
How Much Over-parameterization Is Sufficient to Learn Deep ReLU Networks?
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Distributional Sliced-Wasserstein and Applications to Generative Modeling
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Stochastic Security: Adversarial Defense Using Long-Run Dynamics of Energy-Based Models
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT) @ Virtual
Neural Pruning via Growing Regularization
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Isotropy in the Contextual Embedding Space: Clusters and Manifolds
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Learning perturbation sets for robust machine learning
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Self-supervised Learning from a Multi-view Perspective
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
Estimating and Evaluating Regression Predictive Uncertainty in Deep Object Detectors
[ Paper ]
Poster
Thu May 06 05:00 PM -- 07:00 PM (PDT)
No MCMC for me: Amortized sampling for fast and stable training of energy-based models
[ Paper ]
Oral
Thu May 06 07:00 PM -- 07:15 PM (PDT)
Theoretical Analysis of Self-Training with Deep Networks on Unlabeled Data
[ Paper ]
Spotlight
Thu May 06 07:15 PM -- 07:25 PM (PDT)
Long-tailed Recognition by Routing Diverse Distribution-Aware Experts
[ Paper ]
Spotlight
Thu May 06 07:25 PM -- 07:35 PM (PDT)
Self-Supervised Policy Adaptation during Deployment
[ Paper ]
Spotlight
Thu May 06 07:35 PM -- 07:45 PM (PDT)
What are the Statistical Limits of Offline RL with Linear Function Approximation?
[ Paper ]
Spotlight
Thu May 06 07:55 PM -- 08:05 PM (PDT)
RMSprop converges with proper hyper-parameter
[ Paper ]
Spotlight
Thu May 06 08:05 PM -- 08:15 PM (PDT)
A Good Image Generator Is What You Need for High-Resolution Video Synthesis
[ Paper ]
Spotlight
Thu May 06 08:25 PM -- 08:35 PM (PDT)
Learning with Feature-Dependent Label Noise: A Progressive Approach
[ Paper ]
Spotlight
Thu May 06 08:58 PM -- 09:08 PM (PDT)
Learning a Latent Simplex in Input Sparsity Time
[ Paper ]
Spotlight
Thu May 06 09:08 PM -- 09:18 PM (PDT)
Topology-Aware Segmentation Using Discrete Morse Theory
[ Paper ]
Spotlight
Thu May 06 09:18 PM -- 09:28 PM (PDT)
MARS: Markov Molecular Sampling for Multi-objective Drug Discovery
[ Paper ]
Spotlight
Thu May 06 09:28 PM -- 09:38 PM (PDT)
Distributional Sliced-Wasserstein and Applications to Generative Modeling
[ Paper ]
Break
Thu May 06 10:00 PM -- 10:00 PM (PDT)
BREAK:
Remarks
Thu May 06 10:00 PM -- 11:00 PM (PDT)
Closing Remarks
Workshop
Fri May 07 02:30 AM -- 11:30 AM (PDT)
Science and Engineering of Deep Learning
Workshop
Fri May 07 03:30 AM -- 01:30 PM (PDT)
Neural Compression: From Information Theory to Applications
Workshop
Fri May 07 04:45 AM -- 12:00 PM (PDT)
Hardware-Aware Efficient Training of Deep Learning Models
Workshop
Fri May 07 05:00 AM -- 04:00 PM (PDT)
Geometric and Topological Representation Learning
Workshop
Fri May 07 05:00 AM -- 03:00 PM (PDT)
S2D-OLAD: From shallow to deep, overcoming limited and adverse data
Workshop
Fri May 07 05:15 AM -- 10:00 AM (PDT)
Beyond Static Papers: Rethinking How We Share Scientific Understanding in ML
Workshop
Fri May 07 05:45 AM -- 04:30 PM (PDT)
Self-Supervision for Reinforcement Learning
Workshop
Fri May 07 05:50 AM -- 04:10 PM (PDT)
Energy-Based Models: Current Perspectives, Challenges, and Opportunities
Workshop
Fri May 07 05:55 AM -- 03:00 PM (PDT)
AI for Public Health
Workshop
Fri May 07 05:55 AM -- 02:35 PM (PDT)
The Role of Mathematical Reasoning in General Artificial Intelligence
Break
Fri May 07 06:00 AM -- 06:00 AM (PDT)
BREAK:
Workshop
Fri May 07 06:00 AM -- 01:55 PM (PDT)
A Roadmap to Never-Ending RL
Workshop
Fri May 07 06:00 AM -- 03:00 PM (PDT)
Workshop on Neural Architecture Search
Workshop
Fri May 07 06:00 AM -- 10:00 AM (PDT)
AIMOCC -- AI: Modeling Oceans and Climate Change
Workshop
Fri May 07 06:30 AM -- 05:30 PM (PDT)
How Can Findings About The Brain Improve AI Systems?
Workshop
Fri May 07 06:45 AM -- 07:00 PM (PDT)
Responsible AI (RAI)
Workshop
Fri May 07 07:00 AM -- 02:30 PM (PDT)
Synthetic Data Generation: Quality, Privacy, Bias
Workshop
Fri May 07 07:00 AM -- 12:45 PM (PDT)
Workshop on Learning to Learn
Workshop
Fri May 07 07:00 AM -- 08:30 PM (PDT)
Neural Conversational AI: Bridging the Gap Between Research and Real World (NeuCAIR)
Workshop
Fri May 07 07:00 AM -- 06:00 PM (PDT)
Workshop on Weakly Supervised Learning
Workshop
Fri May 07 07:00 AM -- 12:15 PM (PDT)
2nd Workshop on Practical ML for Developing Countries: Learning Under Limited/low Resource Scenarios
Workshop
Fri May 07 07:00 AM -- 03:55 PM (PDT)
Generalization beyond the training distribution in brains and machines
Workshop
Fri May 07 07:45 AM -- 05:00 PM (PDT)
Workshop on Enormous Language Models: Perspectives and Benchmarks
Workshop
Fri May 07 07:55 AM -- 03:35 PM (PDT)
ICLR 2021 Workshop on Embodied Multimodal Learning (EML)
Workshop
Fri May 07 08:00 AM -- 05:05 PM (PDT)
Robust and reliable machine learning in the real world
Workshop
Fri May 07 08:30 AM -- 01:00 PM (PDT)
Workshop on Distributed and Private Machine Learning
Workshop
Fri May 07 08:45 AM -- 05:00 PM (PDT)
Deep Learning for Simulation
Workshop
Fri May 07 08:45 AM -- 05:50 PM (PDT)
Machine Learning for Preventing and Combating Pandemics
Workshop
Fri May 07 08:45 AM -- 06:00 PM (PDT)
Security and Safety in Machine Learning Systems
Break
Fri May 07 02:00 PM -- 02:00 PM (PDT)
BREAK: