In this talk, we will try to clarify different levels and mechanisms of intelligence from historical, scientific, mathematical, and computational perspective. From the evolution of intelligence in nature, from phylogenetic, to ontogenetic, to societal, and to artificial intelligence, we will try to shed light on how to understand the true nature of the seemingly dramatic advancements in the technologies of machine intelligence in the past decade. We achieve this goal by developing a principled mathematical framework to explain the practice of deep learning from the perspective of compressive data encoding and decoding. This framework not only reveals true nature hence limitations of the current practice and but also provides principled guidelines to develop more complete and more efficient learning architectures and systems. Eventually, we will clarify the difference and relationship between Knowledge and Intelligence, which may guide us to pursue the goal of developing systems with true intelligence.
In this talk, we will try to clarify different levels and mechanisms of intelligence from historical, scientific, mathematical, and computational perspective. From the evolution of intelligence in nature, from phylogenetic, to ontogenetic, to societal, and to artificial intelligence, we will try to shed light on how to understand the true nature of the seemingly dramatic advancements in the technologies of machine intelligence in the past decade. We achieve this goal by developing a principled mathematical framework to explain the practice of deep learning from the perspective of compressive data encoding and decoding. This framework not only reveals true nature hence limitations of the current practice and but also provides principled guidelines to develop more complete and more efficient learning architectures and systems. Eventually, we will clarify the difference and relationship between Knowledge and Intelligence, which may guide us to pursue the goal of developing systems with true intelligence.
Women in Machine Learning
WiML is at ICLR! 🌟
➡️ Luma registration (not required)
Women in Machine Learning (WiML) hosts its signature social at ICLR 2025 - grab lunch, meet fellow researchers, and hear perspectives on navigating academia vs. industry.
🍽️ Lunch provided!
SCHEDULE:
12:30 - 12:35 PM: Opening Remarks
- Melis Ilayda Bal (MPI-IS), Vasiliki Tassopoulou (U Penn)
12:35 - 1:20 PM: Networking & Lunch
- Icebreaker game (12:35 – 12:55 PM)
- Lunch is served (up to capacity; \~1 PM)
- Roundtable discussions (12:55 – 1:20 PM)
1:20 - 2:00 PM: Panel Discussion
"Papers, patents, or products? Making the right career call across academia & industry"
The panel explores key career decisions in today's ML landscape: choosing between research publications and product development, weighing academic freedom against industry resources.
Panelists:
- Reyhane Askari (FAIR)
- Katherine Driscoll (Graph Therapeutics)
- Nouha Dziri (AI2)
- Claire Vernade (University of Tübingen)
Moderator: Erin Grant (WiML; UCL)
Mentorship Hour
MENTORS: Samy Bengio, Xuezhi Wang, Fei Liu
Part of the ICLR experience is meeting people and talking with them about their research interests and experiences. To facilitate these conversations, we are thrilled to announce the third iteration of Mentoring Chats at ICLR (previously called Office Hours).Mentoring Chats will be 45-minute round-table sessions, held during lunch (12:30-1:15 pm and 1:15-2:00 pm) in the Topaz Concourse every day of the main conference (April 24-26). There will be a bell ring approximately 22 minutes in, urging participants to switch tables, or switch topics while staying at the same table.Following ICLR 2024, we have a list of topics and questions that you may wish to ask mentors. We hope to see you there!
Research agenda
- Where should I start if I want to do research in ML? What kind of mathematical/programming skills are required for ML research?
- What are good courses to take? How should I use different modes of learning, such as classroom courses, video lectures, and reading a book?
- How to keep track of all the research literature? How to balance breadth vs depth?
- What are some broader goals of academic machine learning research in the era of LLMs?
- How can one set themselves apart in this crowded research space?
- What is ethical research?
- How to decide on a research area? How to decide on a research project?
- How to adapt my research according to the current trends/community interests?
- How to cope with the pressure of publishing while working on riskier/harder projects? Should I be worried about other groups scooping my research and how to deal with such situations?
- Should I establish myself as an expert in one area/technique or explore a breadth of topics? Should I master a technique and apply it to different problems, or should I master a subfield by finding all useful techniques (hammer vs nails)?
ML+X: Multidisciplinary research
- What are good strategies for starting an interdisciplinary project?
- When working across disciplines, should I have one of them as my “home” community or try to be equally visible in both?
- What are the most efficient ways to help establish my ML+X area as a more active area? Should I organize workshops, teach tutorials, ...?
- How to deal with different incentive structures in interdisciplinary collaborations (e.g., journals vs conferences)?
Advisor and collaborators
- Should I follow my advisor’s agenda or define my own?
- What are the pros and cons of being co-advised?
- When is it appropriate to change advisors and how to go about it?
- How to navigate conflicts with an advisor?
- How to get a good balance between collaborating with other researchers while also distinguishing my own research? Will too much collaboration hurt my job prospects?
- What to look for in a collaborator?
- How do I convey the level of commitment I am willing to have in a project without it being awkward? How to say no to collaborations?
- How to navigate different conventions wrt author ordering? Alphabetical vs contributional ordering? Should my advisor always be a coauthor because they are funding me?
- What do I do if my collaborator is not responsive?
Communicating research and networking
- How to find mentors and allies beyond my advisor?
- What is the best way to communicate my research? Blogs, videos, presentations?
- How to write a good research statement? How to apply for fellowships?
- Should I present my work in poster sessions and workshops? Should I be scared of getting scooped? What are the pros of presenting my work early?
Beyond your institution: Internships and research visits
- Should I do a research internship on a topic different from my dissertation?
- Does it make sense to do a software engineering/development internship if it is not research-related?
- When is a good time to look for internships? Should I apply online or email people?
- Should I do research visits to other universities? Does it make sense to go to semester-long programs as a junior student?
- How to get the most out of my internship? What should be the main goal of doing an internship?
Planning after grad school: academia vs industry
- What should I consider when planning for the next step? How should I decide whether to go to academia or industry?
- How to select a postdoc advisor?
- Should I apply to different departments than my core department? How can I prepare for that, and how early?
- Is it ok to quit your PhD? How can I plan my next steps if so?
Work ethics, open research discussion, personal challenges
- How to balance work-life? How much work is too much work?
- How to take care of mental and physical health?
- How to learn about the ethical implications around the topics of my research?
- How to foster inclusion in research and teaching?
As AI systems become more complex and widely deployed, ensuring their safety is more critical than ever. This social session invites ICRL participants to explore practical approaches to AI safety. Through case studies and interactive discussions, we will delve into methodologies such as red teaming, adversarial testing, guardrails, and human alignment. These examples will also explore how cultural and linguistic diversity influences model evaluations and safety considerations. Whether you’re an AI researcher, engineer, or simply passionate about responsible AI, this session offers a chance to connect, exchange ideas, and help shape the future of safer AI systems.
AI Safety
This social is for people who are interested in or working on AI safety to gather together to discuss various topics around AI safety.
Topics include (not exhaustive): AI alignment: Doing technical and conceptual research focused on getting AI systems to do what we want them to do. AI policy and governance: Setting up institutions and mechanisms that cause major actors (such as AI companies and national governments) to implement good AI safety practices. AI strategy and forecasting: Building models of how AI will develop and which actions can make it go better.
AutoGluon 1.2: Advancing AutoML with Foundation Models and LLM Agents
Automated Machine Learning (AutoML) continues to revolutionize how machine learning models are developed, making it accessible to practitioners with varying levels of expertise. In this workshop, we present the latest advancements in AutoGluon 1.2, an open-source AutoML toolkit developed by Amazon, which empowers users to achieve state-of-the-art performance across diverse machine learning tasks with minimal coding effort. We will emphasize how foundational models can streamline and enhance AutoML performance. Specifically, we will discuss our TabPFN-Mix and Chronos foundational model families for tabular and time series data, respectively. In addition, we will introduce the real-world problems that AutoGluon can help you solve within three lines of code and the fundamental techniques adopted in the toolkit. Rather than diving deep into the mechanisms underlining each individual ML model, we emphasize on how you can take advantage of a diverse collection of models to build an automated ML pipeline.
Human Attention is NOT all you need
While attention does seem to be all you need to attain high-performing models, spanning many modalities, human attention is not all you need to prepare datasets at scale. What you need is scalable and flexible data workflows to automate the process. Established solutions to, e.g., computer vision, audio, and text — including the latest advancements in foundation model capabilities — open up new possibilities to transform AI Data workflows. To name but a few automations, high-volume manual actions like cropping, transcription, and audio-video pairing, as well as more complex reasoning tasks such as video insight extraction and content evaluation. By chaining multiple models together, teams can build custom data engines to create novel, high-quality datasets at scale. On a fixed 100hour human labor budget, we showcase how a high level of automation and constrained budget of human attention spent wisely, consistently outperforms the traditional methods for building datasets.
Bridging Specialized ML Research and Systematic Investing: Transforming the Research Pipeline
Recent breakthroughs in machine learning offer powerful tools that can significantly transform systematic investment research. Within our quantitative research and development team, we leverage specialized ML techniques—including large language models (LLMs), retrieval-augmented generation (RAG), agent-based systems, variational autoencoders (VAEs), graph neural networks (GNNs), and multimodal signal processing—to curate large-scale datasets, automate feature extraction, construct robust trading signals, and systematically generate innovative investment hypotheses.
This talk will identify key opportunities where advanced ML methods, featured in ICLR 2025 papers, can substantially enhance systematic investment pipelines. Additionally, we propose ambitious directions for future research, such as building sophisticated ecosystems of interacting ML agents, creating a compelling landscape for ML researchers interested in translating research innovation into impactful real-world investment strategies.
ML for Accelerating Scientific Discovery: Challenges and Opportunities
Data-driven learning techniques like deep learning (DL) are becoming ubiquitous in various scientific disciplines like computational fluid dynamics, materials science, biological sciences, cyber-physical systems and other science and engineering disciplines. Most often DL techniques, (due to their ability to capture highly non-linear relationships) are employed as `cheap' surrogates to expensive computational simulations or real-world experiments. However, certain characteristic behaviors of DL models like their data-hungry nature, spectral bias, rollout error and lack of explainable decision-making often limit their effectiveness in scientific disciplines. This social will serve as a forum to highlight these technical challenges while also discussing a few potential solutions to better leverage data-driven techniques to further accelerate scientific discovery.
EUREKA: Evaluating and Understanding Large Foundation Models
Rigorous evaluation of large foundation models is critical for assessing the state of the art, informing improvements, and guiding scientific advances in AI. It’s also crucial for app developers using these models. However, practical challenges include benchmark saturation, lack of transparency, difficulties in measuring generative tasks, and numerous capabilities needed for comprehensive model comparison. We also need a deeper understanding of model failures and whether they are consistent over time.
Moreover, with models advancing in reasoning capabilities, a robust evaluation framework is necessary. This session introduces Eureka as a reusable and open framework for standardizing evaluations beyond single-score reporting. We’ll also present Eureka-Bench, which offers benchmarks for challenging and fundamental capabilities in language and vision, including reasoning skills (math, science, hard algorithmic and planning problems). Non-saturated benchmarks help identify meaningful differences between models.
We’ll present insights from analyzing 12 state-of-the-art models, uncovering granular weaknesses and guiding targeted improvements. We’ll also highlight findings from our recent paper on inference-time scaling, which examines reasoning performance and compute tradeoffs. We present an empirical study of inference-time scaling methods for improving reasoning in LLMs across diverse, complex tasks, analyzing their effectiveness, cost-efficiency, and limitations.
Eureka, available as open-source, fosters transparent and reproducible evaluations and has gained significant industry interest, including in prominent press releases.
Useful links:
Blog: https://aka.ms/eureka-ml-insights-blog
Technical report on Eureka: https://aka.ms/eureka-ml-insights-report
Paper on Inference Time Scaling: https://arxiv.org/abs/2504.00294v1
Github repository: https://github.com/microsoft/eureka-ml-insights
Website: https://microsoft.github.io/eureka-ml-insights
Mentorship Hour
MENTORS: Kyunghyun Cho, Aditi Ragunathan, Masashi Sugiyama
Part of the ICLR experience is meeting people and talking with them about their research interests and experiences. To facilitate these conversations, we are thrilled to announce the third iteration of Mentoring Chats at ICLR (previously called Office Hours).Mentoring Chats will be 45-minute round-table sessions, held during lunch (12:30-1:15 pm and 1:15-2:00 pm) in the Topaz Concourse every day of the main conference (April 24-26). There will be a bell ring approximately 22 minutes in, urging participants to switch tables, or switch topics while staying at the same table.Following ICLR 2024, we have a list of topics and questions that you may wish to ask mentors. We hope to see you there!
Research agenda
- Where should I start if I want to do research in ML? What kind of mathematical/programming skills are required for ML research?
- What are good courses to take? How should I use different modes of learning, such as classroom courses, video lectures, and reading a book?
- How to keep track of all the research literature? How to balance breadth vs depth?
- What are some broader goals of academic machine learning research in the era of LLMs?
- How can one set themselves apart in this crowded research space?
- What is ethical research?
- How to decide on a research area? How to decide on a research project?
- How to adapt my research according to the current trends/community interests?
- How to cope with the pressure of publishing while working on riskier/harder projects? Should I be worried about other groups scooping my research and how to deal with such situations?
- Should I establish myself as an expert in one area/technique or explore a breadth of topics? Should I master a technique and apply it to different problems, or should I master a subfield by finding all useful techniques (hammer vs nails)?
ML+X: Multidisciplinary research
- What are good strategies for starting an interdisciplinary project?
- When working across disciplines, should I have one of them as my “home” community or try to be equally visible in both?
- What are the most efficient ways to help establish my ML+X area as a more active area? Should I organize workshops, teach tutorials, ...?
- How to deal with different incentive structures in interdisciplinary collaborations (e.g., journals vs conferences)?
Advisor and collaborators
- Should I follow my advisor’s agenda or define my own?
- What are the pros and cons of being co-advised?
- When is it appropriate to change advisors and how to go about it?
- How to navigate conflicts with an advisor?
- How to get a good balance between collaborating with other researchers while also distinguishing my own research? Will too much collaboration hurt my job prospects?
- What to look for in a collaborator?
- How do I convey the level of commitment I am willing to have in a project without it being awkward? How to say no to collaborations?
- How to navigate different conventions wrt author ordering? Alphabetical vs contributional ordering? Should my advisor always be a coauthor because they are funding me?
- What do I do if my collaborator is not responsive?
Communicating research and networking
- How to find mentors and allies beyond my advisor?
- What is the best way to communicate my research? Blogs, videos, presentations?
- How to write a good research statement? How to apply for fellowships?
- Should I present my work in poster sessions and workshops? Should I be scared of getting scooped? What are the pros of presenting my work early?
Beyond your institution: Internships and research visits
- Should I do a research internship on a topic different from my dissertation?
- Does it make sense to do a software engineering/development internship if it is not research-related?
- When is a good time to look for internships? Should I apply online or email people?
- Should I do research visits to other universities? Does it make sense to go to semester-long programs as a junior student?
- How to get the most out of my internship? What should be the main goal of doing an internship?
Planning after grad school: academia vs industry
- What should I consider when planning for the next step? How should I decide whether to go to academia or industry?
- How to select a postdoc advisor?
- Should I apply to different departments than my core department? How can I prepare for that, and how early?
- Is it ok to quit your PhD? How can I plan my next steps if so?
Work ethics, open research discussion, personal challenges
- How to balance work-life? How much work is too much work?
- How to take care of mental and physical health?
- How to learn about the ethical implications around the topics of my research?
- How to foster inclusion in research and teaching?
Recent advancements in AI and LLM agents have unlocked powerful new capabilities across a wide range of applications. However, these advancements also bring significant risks that must be addressed. In this talk, I will explore the various risks associated with building and deploying AI and LLM agents and discuss approaches to mitigate them. I will also examine how frontier AI and LLM agents could be misused, particularly in cyber security attacks, and how they may reshape the cyber security landscape. Ensuring a safe AI future demands a sociotechnical approach. I will outline our recent proposal for a science- and evidence-based AI policy, highlighting key priorities to deepen our understanding of AI risks, develop effective mitigation approaches, and guide the development of robust AI policies.
Data-Centric AI Social
To sign up for this social, please register on this Luma page: https://lu.ma/rmyoy2vw
Join folks from teams like Cohere for AI, Data Provenance Initiative and Encord working on data-specific AI problems for a Data-Centric AI social!
Data quality has been one of the biggest drivers of AI advancements to date, from large scale data collection efforts such as ImageNet and Common Crawl, to innovations in targeted data curation such as human feedback and preference gathering for RLHF. A number of data-specific AI challenges have also gained visibility in the past year, in part due to to new opportunities for multimodal data, data mixtures and other data-driven research directions, as well as in response to public backlash surrounding to the reuse of online data for model pretraining, which has raised global discourse around the provenance, governance, and copyright considerations.
This social convenes researchers and practitioners focused on a broad range of data topics, spotlighting key organizations working in this space such as the Data Provenance Initiative and Cohere for AI. Following brief opening remarks, there will be a participant-driven unconference, and finally a mixer. The purpose of this social is to discuss shared challenges and opportunities for further dialogue and collaboration following ICLR.
Open-source Decentralized AI Community Social hosted by Flower
This social/hall would aim to bring people together to discuss the interesting open problems in the domain of decentralized AI (and related fields like private ML, federatd learning etc.) and to look for more opportunities to work together. We would aim to faciliate informal sharing of (1) recent work and results, and (2) proposal of collective action; the overal focus would be on how we can better mobilize to lift the quality and quantity of research in this area. Additional topics of interest include: improved benchmarks, realistic datasets and advancing the common understanding of what are the most important settings for decentralized AI to tackle? The primary hosts would be Flower which is the most popular open-source framework for federated and decentralized AI and academics from the University of Cambridge. If accepted we would also attempt to invite other interested parties to attend to add more life to the proceedings.
Test-Time Scaling for LLMs
Test time scaling for LLMs is an emerging frontier that examines how large language models adapt and evolve their responses though reasoning abilities during inference. In recent years, LLMs have demonstrated significant progress in mimicking human-like reasoning—from basic pattern recognition to advanced problem-solving in mathematical contexts. In our social, titled “Test Time Scaling for LLMs,” we delve into the critical role of reasoning and cognitive thinking in LLMs, charting their evolution from earlier foundational works to the cutting-edge models of today. We invite researchers, practitioners, and enthusiasts to share insights and challenges, fostering a collaborative dialogue geared towards understanding the trade-offs between computational cost and reasoning quality. Moreover, we extend our conversation beyond pure mathematics to consider applications in other domains like medical diagnostics, where improved reasoning can lead to safer and more accurate outcomes. This session aims to inspire new ideas for leveraging test time scaling to drive the next wave of advancements in artificial intelligence.
LatinX in AI Social
The LatinX in AI (LXAI) Social is a community-driven gathering aimed at fostering connections among LatinX and Hispanic researchers, engineers, and professionals in AI and ML. This event provides a welcoming space for attendees to network, exchange ideas, and discuss key topics related to AI research, industry trends, and diversity in the field. While this social will not include paper presentations, we plan to feature a few invited speakers to spark meaningful conversations on the challenges and opportunities for LatinX professionals in AI. Our goal is to strengthen the global LXAI community, providing attendees with the opportunity to build professional relationships and collaborate on future initiatives.