(84 events)
Timezone: »
Show all »
Toggle Poster Visibility
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #1
AdaShift: Decorrelation and Convergence of Adaptive Learning Rate Methods
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #2
Self-Tuning Networks: Bilevel Optimization of Hyperparameters using Structured Best-Response Functions
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #3
G-SGD: Optimizing ReLU Neural Networks in its Positively Scale-Invariant Space
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #4
Learning to Make Analogies by Contrasting Abstract Relational Structure
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #5
Deep Frank-Wolfe For Neural Network Optimization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #6
Off-Policy Evaluation and Learning from Logged Bandit Feedback: Error Reduction via Surrogate Policy
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #7
A2BCD: Asynchronous Acceleration with Optimal Complexity
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #8
An analytic theory of generalization dynamics and transfer learning in deep linear networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #9
Minimal Random Code Learning: Getting Bits Back from Compressed Model Parameters
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #10
ANYTIME MINIBATCH: EXPLOITING STRAGGLERS IN ONLINE DISTRIBUTED OPTIMIZATION
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #11
Towards Understanding Regularization in Batch Normalization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #12
A Mean Field Theory of Batch Normalization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #13
Predicting the Generalization Gap in Deep Networks with Margin Distributions
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #14
An Empirical study of Binary Neural Networks' Optimisation
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #15
Deterministic PAC-Bayesian generalization bounds for deep networks via generalizing noise-resilience
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #16
Efficient Training on Very Large Corpora via Gramian Estimation
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #17
Small nonlinearities in activation functions create bad local minima in neural networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #18
Fluctuation-dissipation relations for stochastic gradient descent
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #20
The Comparative Power of ReLU Networks and Polynomial Kernels in the Presence of Sparse Latent Structure
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #21
Optimal Control Via Neural Networks: A Convex Approach
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #22
NOODL: Provable Online Dictionary Learning and Sparse Coding
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #23
Energy-Constrained Compression for Deep Neural Networks via Weighted Sparse Projection and Layer Input Masking
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #24
SGD Converges to Global Minimum in Deep Learning via Star-convex Path
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #25
Deep Lagrangian Networks: Using Physics as Model Prior for Deep Learning
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #26
Relaxed Quantization for Discretized Neural Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #27
There Are Many Consistent Explanations of Unlabeled Data: Why You Should Average
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #28
signSGD with Majority Vote is Communication Efficient and Fault Tolerant
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #29
Preconditioner on Matrix Lie Group for SGD
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #30
A Closer Look at Deep Learning Heuristics: Learning rate restarts, Warmup and Distillation
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #31
Max-MIG: an Information Theoretic Approach for Joint Learning from Crowds
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #32
Rethinking the Value of Network Pruning
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #33
Learning Embeddings into Entropic Wasserstein Spaces
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #34
Deep Layers as Stochastic Solvers
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #35
Initialized Equilibrium Propagation for Backprop-Free Training
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #36
Caveats for information bottleneck in deterministic scenarios
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #37
Learning Two-layer Neural Networks with Symmetric Inputs
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #38
Sparse Dictionary Learning by Dynamical Neural Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #39
Universal Stagewise Learning for Non-Convex Problems with Convergence on Averaged Solutions
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #40
Gradient descent aligns the layers of deep linear networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #41
Stochastic Gradient/Mirror Descent: Minimax Optimality and Implicit Regularization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #42
Learning Self-Imitating Diverse Policies
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #43
On the Universal Approximability and Complexity Bounds of Quantized ReLU Neural Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #44
Adaptive Gradient Methods with Dynamic Bound of Learning Rate
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #46
Per-Tensor Fixed-Point Quantization of the Back-Propagation Algorithm
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #47
The role of over-parametrization in generalization of neural networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #48
Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #50
Understanding Straight-Through Estimator in Training Activation Quantized Neural Nets
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #51
Learning concise representations for regression by evolving networks of trees
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #52
Optimistic mirror descent in saddle-point problems: Going the extra (gradient) mile
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #53
ACCELERATING NONCONVEX LEARNING VIA REPLICA EXCHANGE LANGEVIN DIFFUSION
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #54
The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #55
Three Mechanisms of Weight Decay Regularization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #56
Double Viterbi: Weight Encoding for High Compression Ratio and Fast On-Chip Reconstruction for Deep Neural Network
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #57
Quasi-hyperbolic momentum and Adam for deep learning
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #58
Towards Robust, Locally Linear Deep Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #59
InfoBot: Transfer and Exploration via the Information Bottleneck
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #61
Aggregated Momentum: Stability Through Passive Damping
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #62
From Hard to Soft: Understanding Deep Network Nonlinearities via Vector Quantization and Statistical Inference
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #63
Riemannian Adaptive Optimization Methods
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #64
Regularized Learning for Domain Adaptation under Label Shifts
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #65
DeepOBS: A Deep Learning Optimizer Benchmark Suite
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #66
Fixup Initialization: Residual Learning Without Normalization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #67
Learning sparse relational transition models
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #68
Gradient Descent Provably Optimizes Over-parameterized Neural Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #69
On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #70
A Kernel Random Matrix-Based Approach for Sparse PCA
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #71
SNIP: SINGLE-SHOT NETWORK PRUNING BASED ON CONNECTION SENSITIVITY
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #72
Critical Learning Periods in Deep Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #73
Local SGD Converges Fast and Communicates Little
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #74
Adaptivity of deep ReLU network for learning in Besov and mixed smooth Besov spaces: optimal rate and curse of dimensionality
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #75
A Convergence Analysis of Gradient Descent for Deep Linear Neural Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #76
Analysis of Quantized Models
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #77
Adaptive Estimators Show Information Compression in Deep Neural Networks
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #78
On the Convergence of A Class of Adam-Type Algorithms for Non-Convex Optimization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #79
Theoretical Analysis of Auto Rate-Tuning by Batch Normalization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #80
Decoupled Weight Decay Regularization
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #81
ALISTA: Analytic Weights Are As Good As Learned Weights in LISTA
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #82
Query-Efficient Hard-label Black-box Attack: An Optimization-based Approach
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #83
Minimum Divergence vs. Maximum Margin: an Empirical Comparison on Seq2Seq Models
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #84
Subgradient Descent Learns Orthogonal Dictionaries
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #85
ProxQuant: Quantized Neural Networks via Proximal Operators
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #86
Systematic Generalization: What Is Required and Can It Be Learned?
[ PDF]
Poster
Wed May 08 02:30 PM -- 04:30 PM (PDT) @ Great Hall BC #87
Deep Anomaly Detection with Outlier Exposure
[ PDF]