Machine learning for discrete optimization: Theoretical foundations
Ellen Vitercik, Stanford University
Many of the most important optimization problems in practice are massive in scale, mathematically complex, and involve numerous unknown parameters. Machine learning offers a powerful way to address these challenges by uncovering hidden structure across problem instances, but integrating predictions into algorithms raises fundamental questions: which architectures align with combinatorial structure, and how can we ensure robustness to error? This talk presents two case studies. First, we show how graph neural networks can approximate the optimal dynamic program for online matching, yielding algorithms that generalize across graph sizes and achieve strong empirical performance. Second, we investigate calibration as a principled interface between machine learning and decision-making, demonstrating through rent-or-buy and job scheduling problems that calibrated predictions yield both theoretical guarantees and practical improvements. This is joint work with Alexandre Hayderi, Amin Saberi, Anders Wikum, and Judy Hanwen Shen.
Ellen Vitercik is an Assistant Professor at Stanford University with a joint appointment between the Management Science & Engineering department and the Computer Science department. Her research interests include machine learning, algorithm design, discrete and combinatorial optimization, and the interface between economics and computation. Before joining Stanford, she spent a year as a Miller Postdoctoral Fellow at UC Berkeley and received a PhD in Computer Science from Carnegie Mellon University. Her research has been recognized with a Schmidt Sciences AI2050 Early Career Fellowship, an NSF CAREER award, the SIGecom Doctoral Dissertation Award, and the CMU School of Computer Science Distinguished Dissertation Award, among other honors.
Self-play Algorithms for Math Theorem Proving
Tengyu Ma, Stanford University
I will discuss RL algorithms for automated theorem proving with LLMs, especially in the possible future regime where we run out of high-quality training data. To keep improving the models with limited data, we draw inspiration from mathematicians, who continuously develop new results, partly by proposing novel conjectures or exercises and attempting to solve them. We design the Self-play Theorem Prover (STP) that simultaneously takes on two roles, conjecturer and prover, each providing training signals to the other. At the end of the talk, I will mention a recent paper on extending the algorithm to include another role, Guide, which helps guide the conjecturer to generate clean and relevant conjectures, and a few other related works in using AI for math.
Tengyu Ma is an assistant professor of computer science at Stanford University. His research interests broadly include topics in machine learning, algorithms and their theory, such as deep learning, (deep) reinforcement learning, pre-training / foundation models, robustness, non-convex optimization, distributed optimization, and high-dimensional statistics.
ComPO: Preference Alignment via Comparison Oracles
Tianyi Lin, Columbia University
Direct alignment methods are increasingly used for aligning large language models (LLMs) with human preferences. However, these methods suffer from the likelihood displacement, which can be driven by noisy preference pairs that induce similar likelihood for preferred and dis-preferred responses. To address this issue, we consider doing derivative-free optimization based on comparison oracles. First, we propose a new preference alignment method via comparison oracles and provide convergence guarantees for its basic mechanism. Second, we improve our method using some heuristics and conduct the experiments to demonstrate the flexibility and compatibility of practical mechanisms in improving the performance of LLMs using noisy preference pairs. Evaluations are conducted across multiple base and instruction-tuned models with different benchmarks. Experimental results show the effectiveness of our method as an alternative to addressing the limitations of existing methods. A highlight of our work is that we evidence the importance of designing specialized methods for preference pairs with distinct likelihood margins.
Tianyi Lin is an assistant professor in the Department of Industrial Engineering and Operations Research (IEOR) at Columbia University. His research interests lie in generative artificial intelligence, optimization for machine learning, game theory, social and economic network, and optimal transport. He obtained his Ph.D. in Electrical Engineering and Computer Science at UC Berkeley, where he was advised by Professor Michael Jordan and was associated with the Berkeley Artificial Intelligence Research (BAIR) group. From 2023 to 2024, he was a postdoctoral researcher at the Laboratory for Information & Decision Systems (LIDS) at Massachusetts Institute of Technology, working with Professor Asuman Ozdaglar. Prior to that, he received a B.S. in Mathematics from Nanjing University, a M.S. in Pure Mathematics and Statistics from University of Cambridge and a M.S. in Operations Research from UC Berkeley.
A non-equilibrium phase transition with broken ergodicity leads to double descent and benign overfitting in machine learning
Nigel Goldenfeld, UC San Diego
The remarkable ability of modern neural networks to generalize improves with increasing network capacity, even when the number of model parameters or effective degrees of freedom exceeds the number of training data points. This phenomenon is all the more surprising given that generalization error diverges when the number of model parameters approaches a critical value from below. Here we use dynamical mean field theory to show, in a simple setting of linear regression, that this so-called “double descent” behavior is the outcome of a phase transition in the stochastic field theory describing the training process. We calculate the critical exponents and scaling function of the double descent phase transition, and show that it is marked by a breakdown of the fluctuation-dissipation theorem associated with broken ergodicity. The corresponding response function has the same functional form as the simple London model of the superconducting transition, with the rigidity of the wave function corresponding to the neural network’s ability to generalize accurately. Our results are distinct from earlier work, because we calculate the time-dependence specifically, not just the equilibrium solutions. This is what enables us to identify the origin of the emergent behavior.
Work performed with Chan Li.
Nigel Goldenfeld holds the Chancellor’s Distinguished Professorship in Physics at UC San Diego, which he joined in Fall 2021 after 36 years at the University of Illinois at Urbana-Champaign (UIUC). Nigel’s research spans condensed matter theory, the theory of living systems, hydrodynamics and non-equilibrium statistical physics.
Nigel received his PhD in theoretical physics from the University of Cambridge (UK) in 1982, and from 1982-1985 was a postdoctoral fellow at the Institute for Theoretical Physics at UC Santa Barbara, where his work on the dynamics of snowflake growth helped launch the modern theory of pattern formation in nature. He joined the condensed matter theory group at the Department of Physics at UIUC in 1985, where his work was instrumental to the discovery of d-wave pairing in high temperature superconductors. Nigel’s interests in biology include microbial ecology, evolution and systems biology. He was a founding member of the Institute for Genomic Biology at UIUC, where he led the Biocomplexity Group and directed the NASA Astrobiology Institute for Universal Biology. During the COVID-19 pandemic, he pivoted from his experience in mathematical modeling of bacteria and viruses to computational epidemiology, advising the Governor of Illinois, and helping devise, set up and run the COVID saliva testing system at UIUC, which provided ~12 hour turnaround of PCR tests to the 50,000 people in the campus community and eventually to over 1700 schools and other institutions in Illinois and beyond. Nigel has served on the editorial boards of several journals, including The Philosophical Transactions of the Royal Society, Physical Biology and the International Journal of Theoretical and Applied Finance. Selected honors include: Alfred P. Sloan Foundation Fellow, University Scholar of the University of Illinois, the Xerox Award for research, the A. Nordsieck award for excellence in graduate teaching and the American Physical Society’s Leo P. Kadanoff Prize 2020. Nigel is a Fellow of the American Physical Society, a Fellow of the American Academy of Arts and Sciences, a Fellow of the Royal Society (UK) and a Member of the US National Academy of Sciences.
Inference-Time Algorithms: A Theoretical Lens on Tractability and Error Propagation
Andrej Risteski, Carnegie Mellon
Modern AI systems are increasingly built by placing trained models inside larger computational loops. Inference-time algorithms are a basic instance of this idea: they use one or more trained models at test time to incorporate new information, exploit pretrained models as priors, and trade computational effort for accuracy, sample quality, or control. Examples include generator-verifier search for reasoning, diffusion models for solving inverse problems, and reward-guided generation. Theoretically, this revisits a classical question from optimization and theoretical computer science: what can be done with access to an oracle? Here, however, the oracles are new and non-standard: they model the capabilities of large pretrained models, making them powerful, but also imperfect because they are learned. This combination leads to new questions about algorithm design and error propagation.
This talk studies two central aspects of this paradigm: computational efficiency and error propagation. The first vignette considers generator-verifier systems, and shows how stochastic backtracking can trade additional computation for accuracy, giving a principled version of test-time scaling even with imperfect learned oracles. The second vignette studies diffusion steering: when can we efficiently bias a pretrained diffusion model toward higher-reward samples while staying close to the original model? We show that tractability depends strongly on both the reward structure and the alignment objective, and that simple primitives—such as sampling from linear tilts—can be surprisingly useful for handling richer reward classes.
Andrej Risteski is an Associate Professor at the Machine Learning Department in Carnegie Mellon University. Prior to that, he was a Norbert Wiener Research Fellow jointly in the Applied Math department and IDSS at MIT. Dr. Risteski received his PhD in the Computer Science Department at Princeton University under the advisement of Sanjeev Arora.
Dr. Risteski’s research interests lie in the intersection of machine learning, statistics, and theoretical computer science, spanning topics like (probabilistic) generative models, algorithmic tools for learning and inference, representation and self-supervised learning, out-of-distribution generalization and applications of neural approaches to natural language processing and scientific domains. The broad goal of his research is principled and mathematical understanding of statistical and algorithmic problems arising in modern machine learning paradigms.
Engineering Interpretable and Faithful AI Systems
René Vidal, University of Pennsylvania
Large Language Models (LLMs) and Vision Language Models (VLMs) have achieved remarkable performance across a wide range of tasks. However, their growing deployment has exposed fundamental limitations in faithfulness, safety, and transparency. In this talk, I will present a unified perspective on addressing these challenges through principled model interventions and interpretable decision-making frameworks. I first introduce Information Pursuit (IP), an interpretable-by-design prediction framework that replaces opaque reasoning with a sequence of informative, user-interpretable queries, yielding concise explanations alongside accurate predictions. I then present Parsimonious Concept Engineering (PaCE), an approach that improves faithfulness and alignment by selectively removing undesirable internal activations, mitigating hallucinations and biased language while preserving linguistic competence. Results across text, vision, and medical tasks illustrate how these ideas advance transparency without sacrificing performance. Together, these contributions point toward a broader direction for building AI systems that are powerful, faithful, and aligned with human values.
René Vidal is the Penn Integrates Knowledge and Rachleff University Professor of Electrical and Systems Engineering and Radiology at the University of Pennsylvania, where he directs the Center for Innovation in Data Engineering and Science (IDEAS) and serves as Co-Chair of Penn AI. He is also an Amazon Scholar, Affiliated Chief Scientist at NORCE, and former Associate Editor-in-Chief of IEEE Transactions on Pattern Analysis and Machine Intelligence. Professor Vidal’s research advances the mathematical foundations of deep learning and trustworthy AI, with broad impact across computer vision and biomedical data science. His contributions have been recognized with major honors, including the IEEE Edward J. McCluskey Technical Achievement Award, the D’Alembert Faculty Award, the J.K. Aggarwal Prize, the ONR Young Investigator Award, the NSF CAREER Award, and best paper awards in machine learning, computer vision, signal processing, control, and medical robotics. He is a Fellow of ACM, AIMBE, IEEE, and IAPR, and a Sloan Fellow.
Autopilots Need Parachutes: Reliability Lessons from LLM-Automated Embedded AI Systems
Roberto Morabito, EURECOM
Embedded AI systems are becoming increasingly complex to develop and maintain, requiring specialized workflows that span data processing, model conversion, optimization, and deployment across heterogeneous hardware platforms. Recently, large language models have emerged as a promising tool to automate parts of this lifecycle. In this talk, I present recent work investigating the use of generative AI models as orchestration agents for embedded machine learning pipelines. Using an automated system that leverages LLMs to generate and iteratively refine software artifacts for embedded platforms, we evaluate the feasibility of automating key stages of the AI lifecycle. Our empirical results reveal both the promise and the limitations of this approach. Generative models can significantly accelerate development workflows. However, they also introduce instability, iterative failure modes, and unpredictable operational costs. I will discuss the main failure patterns observed in practice and outline research directions aimed at improving reliability through hybrid reasoning frameworks and system-level feedback mechanisms.
Roberto Morabito is an Assistant Professor in the Networked Systems group of the Communication Systems Department at EURECOM, France, and a Docent at the University of Helsinki. Before joining EURECOM, he was a Senior Researcher in the Department of Computer Science at the University of Helsinki. Earlier in his career, he spent eight years at Ericsson Research Finland, where he worked on cloud platforms, IoT systems, and cyber-physical systems. He received his PhD in Networking Technology from Aalto University in 2019 and was a postdoctoral researcher at the EDGE Lab, School of Electrical and Computer Engineering, Princeton University. His research lies at the intersection of networked systems, edge computing, and distributed AI, focusing on the design and lifecycle management of AI systems operating under computing and networking resource constraints.
Transformers Learn Generalizable Chain-of-Thought Reasoning via Gradient Descent
Yuejie Chi, Yale University
Transformers have demonstrated remarkable chain-of-thought reasoning capabilities, yet, the underlying mechanisms by which they acquire and extrapolate these capabilities remain limited. This talk presents a theoretical analysis of transformers trained via gradient descent for symbolic reasoning and state tracking tasks with increasing problem complexity. Our analysis reveals the coordination of multi-head attention to solve multiple subtasks in a single autoregressive path, and the bootstrapping of inherently sequential reasoning through recursive self-training curriculum. Our optimization-based guarantees demonstrate that even shallow multi-head transformers, with chain-of-thought, can be trained to effectively solve problems that would otherwise require deeper architectures.
Yuejie Chi is the Charles C. and Dorothea S. Dilley Professor of Statistics and Data Science at Yale University, with a secondary appointment in Computer Science, and a member of the Yale Institute for Foundations of Data Science. Before joining Yale, Dr. Chi was the Sense of Wonder Group Endowed Professor of Electrical and Computer Engineering in AI Systems at Carnegie Melon University, with affiliation in MLD and CyLab. She also spent some time as a visiting researcher at Meta's Fundamental AI Research (FAIR). Dr. Yue's research interests lie in the theoretical and algorithmic foundations of data science, generative AI, reinforcement learning, and signal processing, motivated by applications in scientific and engineering domains. Her current focus is on improving the performance, efficiency and reliability of generative AI and decision making, driven by data-intensive but resource-constrained scenarios.
(De)regularized Wasserstein Gradient Flows via Reproducing Kernels
Bharath Sriperumbudur, Pennsylvania State University
Wasserstein gradient flows have become a popular tool in machine learning with applications in sampling, variational inference, generative modeling, and reinforcement learning, among others. The Wasserstein gradient flow (WGF) involves minimizing a probability functional over the Wasserstein space (by taking into account the intrinsic geometry of the Wasserstein space). In this work, we introduce approximate/regularized Wasserstein gradient flows in two different settings: (a) approximate the probability functional and (b) approximate the Wasserstein geometry. In (a), we consider the probability functional to be chi^2-divergence, whose WGF is difficult to implement. To this end, we propose a (de)-regularization of the Maximum Mean Discrepancy (DrMMD) as an approximation of chi^2-divergence and develop an approximate WGF, which is easy to implement and has applications in generative modeling. On the other hand, in the setting of (b), we use Kullback-Leibler divergence as the probability functional and develop an approximation to the Wassertein geometry, which allows for an efficient implementation than that of the exact WGF, with applications in sampling. In both settings, we present a variety of theoretical results that relate the approximate flow to the exact flow and demonstrate the superiority of the approximate flows via numerical simulations.
Bharath Sriperumbudur is a professor in the Department of Statistics (with a courtesy appointment in the Department of Mathematics) at the Pennsylvania State University. His research interests include non-parametric statistics, machine learning, statistical learning theory, optimal transport and gradient flows, regularization and inverse problems, reproducing kernel spaces in probability and statistics, functional and topological data analysis.
Neuromorphic LLMs
Jason Eshraghian, UC Santa Cruz
This talk will show you what neuromorphic computing can do when an academic lab accidentally pulls $2-million of GPU-hours. We will showcase a series of frontier reasoning LLMs developed out of an academic lab, from data curation and pre-training to post-training and alignment. These models surpass leading LLMs from Meta, Google, and other heavily-resourced labs in the ~10-billion parameter regime, despite being 5x smaller.
We have deployed several models on neuromorphic hardware at just 2 watts, bringing state-of-the-art reasoning from the datacenter to the edge. Along the way, we dispel a series of widely-held assumptions about large-scale neuromorphic computation, revealing how it fundamentally differs from conventional deep learning, and why that difference matters.
Jason Eshraghian is an Assistant Professor and Fulbright Scholar in the Department of Electrical and Computer Engineering at the University of California, Santa Cruz. He is the developer of snnTorch, a Python library with over 500,000 downloads for training spiking neural networks. He is a dual-appointed IEEE CAS and EMBS Distinguished Lecturer, an Associate Editor of APL Machine Learning, the Chair of the IEEE Neural Systems and Applications Technical Committee, has been the recipient of seven IEEE Best Paper Awards, a Scientific Advisory Board Member of BrainChip and leads the Neuromorphic Agents Team at Conscium.
Incentivizing Emergent Behaviors for LLMs via Reinforcement Learning
Yi Wu, Tsinghua University
Reinforcement Learning (RL) has become a powerful post-training method for eliciting advanced behaviors in large language models (LLMs). This talk presents recent results showing how RL can incentivize the emergence of LLM capabilities across three domains: (1) multi-player deduction game, Werewolf, where RL-trained LLM agents develop strategic behaviors and outperform strong human players; (2) agentic search, where large-scale RL enables a 32B model to run multi-step search to answer non-trivial questions beyond commercial baselines; and (3) efficient reasoning, where RL mitigates over-thinking and improves both reliability and compute efficiency.
The papers can be found at
- Werewolf: https://arxiv.org/abs/2310.
18940 (ICML24), https://arxiv.org/abs/2502. 04686 (ICML25) - ASearcher: https://arxiv.org/abs/2508.
07976 - Thinking Efficiency: https://www.arxiv.org/abs/
2506.07104 (NeurIPS25)
All the projects are trained using our large-scale agentic RL system, AReaL, which is open-source at https://github.com/
Yi Wu is an assistant professor at the Institute for Interdisciplinary Information Sciences (IIIS), Tsinghua University. He obtained his Ph.D. from UC Berkeley and was a researcher at OpenAI from 2019 to 2020. His research focuses on reinforcement learning, multi-agent learning, and LLM agents. His representative works include the value iteration network, the MADDPG/MAPPO algorithm, OpenAI’s hide-and-seek project, and the AReaL project. He received the best paper award at NIPS 2016, the best demo award finalist at ICRA 2024, and MIT TR35 Asia Pacific 2025 award.
Stochastic-Gradient and Diagonal-Scaling Algorithms for Constrained Optimization and Learning
Frank E. Curtis, Lehigh University
I will motivate and provide an overview of recent efforts in my research group on the design and analysis of stochastic-gradient-based algorithms for solving constrained optimization problems. I will focus in particular on our motivation for informed supervised learning, where constraints in the training problem can be used to impose prior knowledge on the properties that should be possessed by a trained prediction model. In addition, I will provide a detailed look at our newest extensions of heavy-ball and Adam schemes from the unconstrained to the equality-constrained setting, for which we have shown state-of-the-art convergence guarantees. I will demonstrate the impressive practical performance of our methods using a few informed supervised learning problems.
Frank E. Curtis is a Professor in the Department of Industrial and Systems Engineering at Lehigh University, where he has been employed since 2009. He received a bachelor’s degree from the College of William and Mary in 2003 with a double major in Computer Science and Mathematics, received a master’s degree in 2004 and Ph.D. degree in 2007 from the Department of Industrial Engineering and Management Science at Northwestern University, and spent two years as a Postdoctoral Researcher in the Courant Institute of Mathematical Sciences at New York University from 2007 until 2009. His research focuses on the design, analysis, and implementation of numerical methods for solving large-scale nonlinear optimization problems. He received an Early Career Award from the Advanced Scientific Computing Research (ASCR) program of the U.S. Department of Energy (DoE), and has received funding from various programs of the U.S. National Science Foundation (NSF), including through a TRIPODS Phase I grant awarded to him and his collaborators at Lehigh, Northwestern, and Boston University. He has also received funding from the U.S. Office of Naval Research (ONR) and DoE’s Advanced Research Projects Agency-Energy (ARPA-E). He received, along with Leon Bottou (Meta AI) and Jorge Nocedal (Northwestern), the 2021 SIAM/MOS Lagrange Prize in Continuous Optimization. He was awarded, with James V. Burke (U. of Washington), Adrian Lewis (Cornell), and Michael Overton (NYU), the 2018 INFORMS Computing Society Prize. He and team members Daniel Molzahn (Georgia Tech), Andreas Waechter (Northwestern), Ermin Wei (Northwestern), and Elizabeth Wong (UC San Diego) were awarded second place in the ARPA-E Grid Optimization Competition in 2020. He currently serves as Area Editor for Continuous Optimization for Mathematics of Operations Research and serves as an Associate Editor for Mathematical Programming, SIAM Journal on Optimization, Operations Research, IMA Journal of Numerical Analysis, and Mathematical Programming Computation. He previously served as the Vice Chair for Nonlinear Programming for the INFORMS Optimization Society, and is currently very active in professional societies and groups related to mathematical optimization, including INFORMS, the Mathematics Optimization Society, and the SIAM Activity Group on Optimization.
Training Neural Networks at Any Scale
Volkan Cevher, EPFL
At the heart of deep learning’s transformative impact lies the concept of scale–encompassing both data and computational resources, as well as their interaction with neural network architectures. Scale, however, presents critical challenges, such as increased instability during training and prohibitively expensive model-specific tuning. Given the substantial resources required to train such models, formulating high-confidence scaling hypotheses backed by rigorous theoretical research has become paramount.
To bridge theory and practice, the talk explores a key mathematical ingredient of scaling in tandem with scaling theory: the numerical solution algorithms commonly employed in deep learning, spanning domains from vision to language models. We unify these algorithms under a common master template, making their foundational principles transparent. In doing so, we reveal the interplay between adaptation to smoothness structures via online learning and the exploitation of optimization geometry through non-Euclidean norms. Our exposition moves beyond simply building larger models–it emphasizes strategic scaling, offering insights that promise to advance the field while economizing on resources.
Volkan Cevher received the B.Sc. (valedictorian) in electrical engineering from Bilkent University in Ankara, Turkey, in 1999 and the Ph.D. in electrical and computer engineering from the Georgia Institute of Technology in Atlanta, GA in 2005. He was a Research Scientist with the University of Maryland, College Park from 2006-2007 and also with Rice University in Houston, TX, from 2008-2009. Currently, he is an Associate Professor at the Swiss Federal Institute of Technology Lausanne and a Faculty Fellow in the Electrical and Computer Engineering Department at Rice University. His research interests include machine learning, signal processing theory, optimization theory and methods, and information theory. Dr. Cevher is an ELLIS fellow and was the recipient of the Google Faculty Research award in 2018, the IEEE Signal Processing Society Best Paper Award in 2016, a Best Paper Award at CAMSAP in 2015, a Best Paper Award at SPARS in 2009, and an ERC CG in 2016 as well as an ERC StG in 2011.
Certifiably Correct Machine Perception
David Rosen, Northeastern University
Many fundamental machine perception and state estimation tasks require the solution of a high-dimensional nonconvex estimation problem; this class includes (for example) the fundamental problems of simultaneous localization and mapping (in robotics), 3D reconstruction (in computer vision), and sensor network localization (in distributed sensing). Such problems are known to be computationally hard in general, with many local minima that can entrap the smooth local optimization methods commonly applied to solve them. The result is that standard machine perception algorithms (based upon local optimization) can be surprisingly brittle, often returning egregiously wrong answers even when the problem to which they are applied is well-posed.
In this talk, we present a novel class of certifiably correct estimation algorithms that are capable of efficiently recovering provably good (often globally optimal) solutions of generally-intractable machine perception problems in many practical settings. Our approach directly tackles the problem of nonconvexity by employing convex relaxations whose minimizers provide provably good approximate solutions to the original estimation problem under moderate measurement noise. We illustrate the design of this class of methods using the fundamental problem of pose-graph optimization (a mathematical abstraction of robotic mapping) as a running example. We conclude with a brief discussion of open questions and future research directions.
David M. Rosen is an Assistant Professor in the Departments of Electrical & Computer Engineering and Mathematics and the Khoury College of Computer Sciences (by courtesy) at Northeastern University, where he leads the Robust Autonomy Laboratory (NEURAL). Prior to joining Northeastern, he was a Research Scientist at Oculus Research (now Meta Reality Labs) from 2016 to 2018, and a Postdoctoral Associate at MIT’s Laboratory for Information and Decision Systems (LIDS) from 2018 to 2021. He holds the degrees of B.S. in Mathematics from the California Institute of Technology (2008), M.A. in Mathematics from the University of Texas at Austin (2010), and ScD in Computer Science from the Massachusetts Institute of Technology (2016).
He is broadly interested in the mathematical and algorithmic foundations of trustworthy machine perception, learning, and control. His work has been recognized with the IEEE Transactions on Robotics Best Paper Award (2024), an Honorable Mention for the IEEE Transactions on Robotics Best Paper Award (2021), a Best Student Paper Award at Robotics: Science and Systems (2020), a Best Paper Award at the International Workshop on the Algorithmic Foundations of Robotics (2016), and selection as an RSS Pioneer (2019).
High-dimensional Optimization with Applications to Compute-Optimal Neural Scaling Laws
Courtney Paquette (McGill University)
Given the massive scale of modern ML models, we now only get a single shot to train them effectively. This restricts our ability to test multiple architectures and hyper-parameter configurations. Instead, we need to understand how these models scale, allowing us to experiment with smaller problems and then apply those insights to larger-scale models. In this talk, I will present a framework for analyzing scaling laws in stochastic learning algorithms using a power-law random features model (PLRF), leveraging high-dimensional probability and random matrix theory. I will then use this scaling law to address the compute-optimal question: How should we choose model size and hyper-parameters to achieve the best possible performance in the most compute-efficient manner? Then using this PLRF model, I will devise a new momentum-based algorithm that (provably) improves the scaling law exponent. Finally, I will present some numerical experiments on LSTMs that show how this new stochastic algorithm can be applied to real data to improve the compute-optimal exponent.
Courtney Paquette is an assistant professor at McGill University in the Mathematics and Statistics department, a CIFAR AI Chair (MILA), and an active member of the Montreal Machine Learning Optimization Group (MTL MLOpt) at MILA. Her research broadly focuses on designing and analyzing algorithms for large-scale optimization problems, motivated by applications in data science, and using techniques that draw from a variety of fields, including probability, complexity theory, and convex and nonsmooth analysis. Dr. Paquette is a lead organizer of the OPT-ML Workshop at NeurIPS since 2020, and a lead organizer (and original creator) of the High-dimensional Learning Dynamics (HiLD) Workshop at ICML.
A New Paradigm for Learning with Distribution Shift
Adam Klivans, UT Austin
We revisit the fundamental problem of learning with distribution shift, where a learner is given labeled samples from training distribution D, unlabeled samples from test distribution D′ and is asked to output a classifier with low test error. The standard approach in this setting is to prove a generalization bound in terms of some notion of distance between D and D′. These distances, however, are difficult to compute, and this has been the main stumbling block for efficient algorithm design over the last two decades.
We sidestep this issue and define a new model called TDS learning, where a learner runs a test on the training set and is allowed to reject if this test detects distribution shift relative to a fixed output classifier. This approach leads to the first set of efficient algorithms for learning with distribution shift that do not take any assumptions on the test distribution. Finally, we discuss how our techniques have recently been used to solve longstanding problems in supervised learning with contamination.
Adam Klivans is a Professor of Computer Science at the University of Texas at Austin and Director of the NSF AI Institute for Foundations of Machine Learning (IFML). His research interests lie in machine learning and theoretical computer science, in particular, Learning Theory, Computational Complexity, Pseudorandomness, Limit Theorems, and Gaussian Space. Dr. Klivans is a recipient of the NSF CAREER Award and serves on the editorial board for the Theory of Computing and Machine Learning Journal.