Talks | ISL Colloquium

Upcoming Talks

Qualia Optimization: Exploring Mathematical Formulations of AI Experience

Philip Thomas • Associate Professor, University of Massachusetts

Thursday, February 19, 2026 at 4:00 PM • Packard 202

View details →

The Arrow-Hurwicz-Uzawa Method: Historical Perspectives and Current Advances

Xiaoming Yuan • Professor, The University of Hong Kong

Thursday, February 12, 2026 at 4:00 PM • Packard 202

View details →

Controllable AI: Control Theory meets Artificial Intelligence

Carmen Amo Alonso • Postdoc, Stanford University

Thursday, February 5, 2026 at 4:00 PM • Packard 202

View details →

Post-training diffusion models: a unified view from stochastic analysis

Renyuan Xu • Assistant Professor, Stanford University

Thursday, January 22, 2026 at 4:00 PM • Packard 202

View details →

From Bits to Rounds: Unifying Information-Theoretic Limits and Real-World Efficiency in Diffusion Language Models

Jiantao Jiao • Assistant Professor, UC Berkeley

Thursday, November 20, 2025 at 4:00 PM • Packard 202

View details →

PEPFlow: A library for the systematic analysis of optimization algorithms

Xin Jiang • Assistant Professor, University of Houston

Thursday, November 13, 2025 at 4:00 PM • Packard 202

View details →

Subspace injections

Joel Tropp • Professor, Caltech

Thursday, November 6, 2025 at 4:00 PM • Hewlett 201

View details →

Simplifying Complex Machine Learning Models with Mathematical Optimization

Rahul Mazumder • Associate Professor, MIT

Thursday, October 23, 2025 at 4:00 PM • Packard 202

View details →

Navigating Underspecified Optimization Problems using Portfolios for Human-AI Augmented Decision-Making

Swati Gupta • Associate Professor, MIT

Thursday, October 16, 2025 at 4:00 PM • Packard 202

View details →

Multi-period mixed-integer quadratic programming

Alper Atamturk • Professor, UC Berkeley

Thursday, October 9, 2025 at 4:00 PM • Packard 202

View details →

Slides

Randomized Algorithms for Solving Linear Systems with Low-Rank Structure

Michał Dereziński • Assistant Professor, University of Michigan

Thursday, October 2, 2025 at 4:00 PM • Packard 202

View details →

Toward Trustworthy AI: Principled and Automated Interpretability in Deep Learning

Lily Weng • Assistant Professor, UCSD

Thursday, July 31, 2025 at 4:00 PM • Packard 202

View details →

Negative Stepsizes Make Gradient-Descent-Ascent Converge

Jason Altschuler • Assistant Professor, University of Pennsylvania

Thursday, May 29, 2025 at 4:00 PM • Packard 202

View details →

Rank Overparameterization: Embracing Nonconvexity in Low-rank Optimization

Richard Y. Zhang • Assistant Professor, University of Illinois Urbana-Champaign

Thursday, March 6, 2025 at 4:00 PM • Packard 202

View details →

Differentially private M-estimation via noisy optimization

Po-Ling Loh • Professor, University of Cambridge

Thursday, February 27, 2025 at 4:00 PM • Packard 202

View details →

Exponentially faster gradient methods in the presence of ravines

Damek Davis • Visiting Associate Professor, University of Pennsylvania

Thursday, November 21, 2024 at 4:00 PM • Packard 202

View details →

Towards Pluralistic Alignment: Foundations for Learning Diverse Human Preferences

Ramya Korlakai Vinayak • Assistant Professor, University of Wisconsin-Madison

Thursday, November 14, 2024 at 4:00 PM • Packard 202

View details →

Control Across Scales and Applications to Traffic Management

Murat Arcak • Professor, UC Berkeley

Thursday, May 30, 2024 at 4:00 PM • Packard 202

View details →

Cardinal-Utility Matching Markets: From Tractability to Intractability ... and Back!

Vijay Vazirani • Distinguished Professor, UC Irvine

Thursday, May 23, 2024 at 4:00 PM • Packard 101

View details →

Challenges & Opportunities for Learning in Strategic Environments

Eric Mazumdar • Assistant Professor, Caltech

Thursday, April 25, 2024 at 4:00 PM • Packard 202

View details →

What makes learning to control easy or hard?

Nikolai Matni • Assistant Professor, University of Pennsylvania

Thursday, April 18, 2024 at 4:00 PM • Packard 202

View details →

Slides

Optimization, Robustness and Attention in Deep Learning: Insights from Random and NTK Feature Models

Marco Mondelli • Assistant Professor, IST Austria

Thursday, April 11, 2024 at 4:00 PM • Packard 202

View details →

Design and Analysis of Decentralized Learning Enabled Multi-Agent Systems

Chinmay Maheshwari • PhD Student, UC Berkeley

Thursday, March 14, 2024 at 4:00 PM • Packard 202

View details →

What Algorithms can Transformers Learn? A Study in Length Generalization

Preetum Nakkiran • Research Scientist, Apple Research

Thursday, March 7, 2024 at 4:00 PM • Packard 202

View details →

From Large to Small Datasets: Size Generalization for Clustering Algorithm Selection

Ellen Vitercik • Assistant Professor, Stanford

Thursday, February 29, 2024 at 4:00 PM • Packard 202

View details →

Solving semirandom planted CSPs via SDP-certificates and spectral sparsification

Venkatesan Guruswami • Professor, UC Berkeley

Thursday, February 22, 2024 at 4:00 PM • Packard 202

View details →

Clustering for causal inference under network interference

Johan Ugander • Associate Professor, Stanford

Thursday, February 15, 2024 at 4:00 PM • Packard 202

View details →

Permissionless Blockchains: Fundamental Trade-offs in Throughput, Latency, and Safety

Dongning Guo • Professor, Northwestern University

Thursday, February 8, 2024 at 4:00 PM • Packard 202

View details →

Asymptotic Learning in Overparameterized Models

Anant Sahai • Professor, UC Berkeley

Thursday, January 25, 2024 at 4:00 PM • Packard 202

View details →

Model Selection And Ensembling When There Are More Parameters Than Data

Michael Mahoney • Professor, ICSI, LBNL, and UC Berkeley

Thursday, December 7, 2023 at 4:00 PM • Packard 202

View details →

The Economics of Automated Market Making and Decentralized Exchanges

Ciamac Moallemi • Professor, Columbia University

Thursday, November 30, 2023 at 4:00 PM • Packard 202

View details →

Efficient Neural Scene Representation, Rendering, and Generation

Gordon Wetzstein • Associate Professor, Stanford

Thursday, November 16, 2023 at 4:00 PM • Packard 202

View details →

Compressed Sensing using Generative Models: Theory and Applications

Ajil Jalal • Postdoc, UC Berkeley

Thursday, November 9, 2023 at 4:00 PM • Packard 202

View details →

See versus Hear - Wireless Digital Traffic Signs

Chai Toh • Senior Fellow, UC Berkeley

Thursday, November 2, 2023 at 4:00 PM • Packard 202

View details →

From Optimization to Sampling, with a touch of Schrödinger Bridge

Qijia Jiang • Postdoc, UC Berkeley

Thursday, October 26, 2023 at 4:00 PM • Packard 202

View details →

CICERO: Human-Level Performance in the Game of Diplomacy by Combining Language Models with Strategic Reasoning

Noam Brown • Research Scientist, OpenAI

Thursday, October 19, 2023 at 4:00 PM • Packard 202

View details →

Toward a grand unified theory of accelerations in optimization and machine learning

Ernest Ryu • Assistant Professor, Seoul National University

Thursday, October 12, 2023 at 4:00 PM • Packard 202

View details →

Low rank approximation for faster optimization

Madeleine Udell • Assistant Professor, Stanford

Thursday, June 1, 2023 at 4:00 PM • Packard 202

View details →

DoG is SGD's best friend: toward tuning-free stochastic optimization

Yair Carmon • Assistant Professor, Tel Aviv University

Wednesday, May 31, 2023 at 2:00 PM • Packard 202

View details →

Mathematical underpinnings of emerging computational imaging inverse problems

Shirin Jalali • Assistant Professor, Rutgers University

Thursday, May 25, 2023 at 4:00 PM • Packard 202

View details →

Automating the analysis and design of iterative optimization algorithms

Laurent Lessard • Associate Professor, Northeastern University

Thursday, May 18, 2023 at 4:00 PM • Packard 202

View details →

Representative and Deliberative Social Choice

Ashish Goel • Professor, Stanford

Thursday, April 27, 2023 at 4:00 PM • Packard 202

View details →

New local differentially private protocols for frequency and mean estimation

Jelani Nelson • Professor, UC Berkeley

Friday, April 21, 2023 at 4:00 PM • Packard 202

View details →

Near-Optimal Algorithms for Imitation Learning

Jiantao Jiao • Assistant Professor, UC Berkeley

Thursday, April 13, 2023 at 2:00 PM • Packard 202

View details →

Breaking the Sample Size Barrier in Reinforcement Learning

Yuxin Chen • Associate Professor, University of Pennsylvania

Thursday, April 6, 2023 at 4:00 PM • Packard 202

View details →

Looking beyond the Worst-Case Adversaries in Machine Learning

Nika Haghtalab • Assistant Professor, UC Berkeley

Thursday, March 23, 2023 at 4:00 PM • Packard 202

View details →

An Information Theory for Out-of-Order Information: Applications in DNA Data Storage and Beyond

Ilan Shomorony • Assistant Professor, UIUC

Thursday, March 16, 2023 at 4:00 PM • Packard 202

View details →

Kernel Thinning and Stein Thinning

Lester Mackey • Principal Researcher, Microsoft Research

Thursday, March 9, 2023 at 4:00 PM • Packard 202

View details →

Slides

Causal Representation Learning – A Proposal

Caroline Uhler • Professor, MIT

Friday, March 3, 2023 at 1:00 PM • Packard 202

View details →

Gromov-Wasserstein Distances: Entropic Regularization, Duality, and Sample Complexity

Ziv Goldfeld • Assistant Professor, Cornell University

Thursday, February 23, 2023 at 4:00 PM • Packard 202

View details →

Adversarial machine learning and instrumental variables for flexible causal modeling

Vasilis Syrgkanis • Assistant Professor, Stanford

Thursday, February 2, 2023 at 4:00 PM • Packard 202

View details →

Learning with Feature Geometry

Lizhong Zheng • Professor, MIT

Thursday, January 26, 2023 at 4:00 PM • Packard 202

View details →

Recent Advances in Non-log-concave and Heavy-tailed Sampling

Krishna Balasubramanian • Assistant Professor, UC Davis

Thursday, January 19, 2023 at 4:00 PM • Packard 202

View details →

Next-Generation Communication/Artificial-Intelligence Challenges

John Cioffi • Professor, Stanford

Thursday, January 12, 2023 at 4:00 PM • Packard 202

View details →

What kind of information is there in single-cell genomics data?

Lior Pachter • Professor, Caltech

Thursday, December 8, 2022 at 4:00 PM • Packard 202

View details →

Fundamental limits of low-rank matrix estimation with diverging aspect ratios

Yuchen Wu • PhD student, Stanford

Thursday, December 1, 2022 at 4:00 PM • Packard 202

View details →

Two Vignettes on Efficient Procedures for Personalized Decision Making

Raaz Dwivedi • Postdoc, Harvard/MIT

Thursday, November 17, 2022 at 4:00 PM • Packard 202

View details →

Recent Advances in Colonel Blotto Games and the Connection to Controls

Jason R. Marden • Professor, UCSB

Thursday, November 10, 2022 at 4:00 PM • Packard 202

View details →

Leveraging 'partial' smoothness for faster convergence in nonsmooth optimization

Damek Davis • Associate Professor, Cornell

Thursday, November 3, 2022 at 4:00 PM • Packard 202

View details →

Towards Instance-Optimal Algorithms for Reinforcement Learning

Kevin Jamieson • Assistant Professor, University of Washington

Thursday, October 27, 2022 at 4:00 PM • Packard 202

View details →

A data-centric view on reliable generalization

Ludwig Schmidt • Assistant Professor, University of Washington

Thursday, October 20, 2022 at 4:00 PM • Packard 202

View details →

Stochastic Algorithms in the Large: Batch Size Saturation, Stepsize Criticality, Generalization Performance, and Exact Dynamics

Courtney Paquette • Assistant Professor, McGill University

Thursday, October 13, 2022 at 4:00 PM • Packard 101

View details →

Computational aspects of learning sparse functions with neural networks

Theodor Misiakiewicz • PhD student, Stanford

Thursday, October 6, 2022 at 4:00 PM • Packard 101

View details →

Attacking the privacy of machine learning models

Nicholas Carlini • Research Scientist, Google Brain

Thursday, September 29, 2022 at 4:00 PM • Packard 101

View details →

Algorithms and computational limits for infinite-horizon general-sum stochastic games

Vidya Muthukumar • Professor, Georgia Tech

Thursday, May 19, 2022 at 4:00 PM • Packard 101

View details →

Beyond the Csiszár-Körner Bound: Best-Possible Wiretap Coding via Obfuscation

Amit Sahai • Professor, UCLA

Thursday, May 12, 2022 at 4:00 PM • Packard 101

View details →

The online convex optimization approach to control

Elad Hazan • Professor, Princeton

Thursday, May 5, 2022 at 4:00 PM • Packard 101

View details →

Two vignettes about interpolation and generalization in overparameterized models

Niladri Chatterji • Postdoc, Stanford

Thursday, April 28, 2022 at 4:00 PM • Packard 101

View details →

Historical thoughts on modern prediction

Ben Recht • Professor, UC Berkeley

Thursday, April 21, 2022 at 4:00 PM • Packard 101

View details →

What is the Statistical Complexity of Reinforcement Learning?

Sham Kakade • Professor, Harvard

Thursday, April 14, 2022 at 1:00 PM • Online (Zoom)

View details →

Palo Alto, We Have a Problem. There Is No Oracle!

Amin Karbasi • Professor, Yale University

Thursday, March 17, 2022 at 4:00 PM • Online (Zoom)

View details →

Towards instance-optimal compression for distributed mean estimation

Ananda Theertha Suresh • Research Scientist, Google Research

Thursday, March 10, 2022 at 4:00 PM • Online (Zoom)

View details →

Adaptivity and Confounding in Multi-Armed Bandit Experiments

Daniel Russo • Professor, Columbia Business School

Thursday, March 3, 2022 at 4:00 PM • Online (Zoom)

View details →

Statistical approaches for mechanistic biological discovery with genomics

Julia Salzman • Professor, Stanford

Thursday, February 24, 2022 at 4:00 PM • Online (Zoom)

View details →

Denoising as a Building Block for Imaging, Inverse Problems, and Machine Learning

Peyman Milanfar • Principal Scientist / Director, Google Research

Thursday, February 10, 2022 at 4:00 PM • Online (Zoom)

View details →

A Continuized View on Nesterov Acceleration for Stochastic Gradient Descent and Randomized Gossip

Raphaël Berthier • Postdoc, EPFL

Thursday, February 3, 2022 at 4:00 PM • Packard 101

View details →

The Foundations of Private Statistical Estimation

Jonathan Ullman • Professor, Northeastern University

Thursday, January 13, 2022 at 4:00 PM • Packard 101

View details →

Contextual Search in the Presence of Irrational Agents

Chara Podimata • Graduate Student, Harvard

Thursday, January 6, 2022 at 4:00 PM • Packard 101

View details →

Connecting Statistical Problems with Different Structures

Guy Bresler • Professor, MIT

Thursday, December 2, 2021 at 4:00 PM • Packard 101

View details →

Optimization in Theory and Practice

Stephen Wright • Professor, University of Wisconsin-Madison

Thursday, November 18, 2021 at 4:00 PM • Packard 101

View details →

Safe and Efficient Reinforcement Learning for Power System Control

Baosen Zhang • Professor, University of Washington

Thursday, November 11, 2021 at 4:00 PM • Packard 101

View details →

Distributed Algorithms for Optimization in Networks

Angelia Nedich • Professor, Arizona State University

Thursday, November 4, 2021 at 4:00 PM • Packard 101

View details →

On the Convergence of Langevin Monte Carlo: The Interplay between Tail Growth and Smoothness

Murat Erdogdu • Assistant Professor, University of Toronto

Thursday, October 28, 2021 at 4:00 PM • Packard 101

View details →

AI for clinical trials and clinical trials for AI

James Zou • Professor, Stanford

Thursday, October 21, 2021 at 4:00 PM • Packard 101

View details →

Experimentation and Decision-Making in Two-Sided Marketplaces: The Impact of Interference / Simple Agent, Complex Environment: Efficient Reinforcement Learning with Agent States

Hannah Li / Shi Dong • PhD Students, Stanford

Thursday, October 14, 2021 at 4:00 PM • Packard 101

View details →

KO codes: Inventing Non-linear Encoding and Decoding for Reliable Wireless Communication via Deep-learning

Ashok Vardhan Makkuva • PhD Student, UIUC

Thursday, October 7, 2021 at 4:00 PM • Packard 101

View details →

Sequential Decision Making: How Much Adaptivity Is Needed Anyways?

Amin Karbasi • Professor, Yale

Thursday, September 30, 2021 at 4:45 PM • Packard 101

View details →

Interpolation Phase Transition in Neural Networks: Memorization and Generalization

Joe Zhong • Postdoc, Stanford

Thursday, September 23, 2021 at 4:30 PM • Packard 101

View details →

Between tractable and intractable problems in reinforcement learning

Csaba Szepesvari • Professor, University of Alberta

Thursday, June 10, 2021 at 4:30 PM • Online (Zoom)

View details →

The Hidden Convex Optimization Landscape of Deep Neural Networks

Mert Pilanci • Professor, Stanford

Thursday, June 3, 2021 at 4:30 PM • Online (Zoom)

View details →

Learning from preferences and labels

Aarti Singh • Professor, Carnegie Mellon University

Thursday, May 27, 2021 at 4:30 PM • Online (Zoom)

View details →

Multi-Scale Methods for Machine Learning

Inderjit S. Dhillon • Professor, UT Austin

Thursday, May 20, 2021 at 4:30 PM • Online (Zoom)

View details →

Interior Point Methods for Nearly Linear Time Algorithms

Aaron Sidford • Professor, Stanford

Thursday, May 6, 2021 at 4:30 PM • Online (Zoom)

View details →

Private Stochastic Convex Optimization

Kunal Talwar • Research Scientist, Apple

Thursday, April 29, 2021 at 4:30 PM • Online (Zoom)

View details →

Recent results in planted assignment problems

Yihong Wu • Professor, Yale

Thursday, April 22, 2021 at 4:30 PM • Online (Zoom)

View details →

Demystfying the Efficiency of Reinforcement Learning: Two Recent Stories

Yuxin Chen • Professor, Princeton

Thursday, April 15, 2021 at 4:30 PM • Online (Zoom)

View details →

Scalable semidefinite programming

Joel Tropp • Professor, Caltech

Thursday, April 8, 2021 at 4:30 PM • Online (Zoom)

View details →

Finding Global Minima via Kernel Approximations

Francis Bach • Researcher, INRIA

Thursday, April 1, 2021 at 10:00 AM • Online (Zoom)

View details →

Approximating cross-validation: guarantees for model assessment and selection

Ashia Wilson • Professor, MIT

Thursday, March 18, 2021 at 4:30 PM • Online (Zoom)

View details →

Robust Learning from Batches -- The Best Things in Life are (Almost) Free

Alon Orlitsky • Professor, UC San Diego

Thursday, March 11, 2021 at 4:30 PM • Online (Zoom)

View details →

Exploiting Symmetries in Inference and Learning

Max Welling • Professor, University of Amsterdam

Thursday, March 4, 2021 at 9:00 AM • Online (Zoom)

View details →

Chasing the Long Tail: What Neural Networks Memorize and Why

Vitaly Feldman • Research Scientist, Apple AI Research

Thursday, February 25, 2021 at 4:30 PM • Online (Zoom)

View details →

Slides

Enabling Fast and Robust Federated Learning

Ramtin Pedarsani • Professor, UC Santa Barbara

Thursday, February 18, 2021 at 4:30 PM • Online (Zoom)

View details →

Data, decisions, and dynamics

Moritz Hardt • Professor, UC Berkeley

Thursday, February 11, 2021 at 4:30 PM • Online (Zoom)

View details →

End-to-end learning for computational microscopy

Laura Waller • Professor, UC Berkeley

Thursday, February 4, 2021 at 4:30 PM • Online (Zoom)

View details →

The Well Tempered Lasso

Yoram Singer • Professor, Princeton University

Thursday, January 28, 2021 at 4:30 PM • Online (Zoom)

View details →

Adaptive Experimental Design for Best Identification and Multiple Testing

Kevin Jamieson • Professor, University of Washington

Thursday, January 21, 2021 at 4:30 PM • Online (Zoom)

View details →

Accelerating Reinforcement Learning in Emerging Wireless IoT Systems and Applications via System Awareness

Nicholas Mastronarde & Jacob Chakareski • Professors, State University of New York at Buffalo & New Jersey Institute of Technology

Thursday, January 14, 2021 at 4:30 PM • Online (Zoom)

View details →

Interference in Experimental Design in Online Platforms

Ramesh Johari • Professor, Stanford

Thursday, November 19, 2020 at 4:30 PM • Online (Zoom)

View details →

A Very Sketchy Talk

David Woodruff • Professor, CMU

Friday, November 13, 2020 at 1:15 PM • Online (Zoom)

View details →

Learning Convolutions from Scratch

Behnam Neyshabur • Sr. Research Scientist, Google

Thursday, November 5, 2020 at 4:30 PM • Online (Zoom)

View details →

Towards Model Agnostic Robustness

Sujay Sanghavi • Professor, UT Austin

Thursday, October 29, 2020 at 4:30 PM • Online (Zoom)

View details →

Competitive Control via Online Optimization

Adam Wierman • Professor, Caltech

Thursday, October 22, 2020 at 4:30 PM • Online (Zoom)

View details →

Strategies for Active Machine Learning

Robert Nowak • Professor, University of Wisconsin-Madison

Thursday, October 15, 2020 at 4:30 PM • Online (Zoom)

View details →

Computational Barriers to Estimation from Low-Degree Polynomials

Alex Wein • Instructor, NYU Courant Institute

Thursday, October 8, 2020 at 4:30 PM • Online (Zoom)

View details →

Optimizing the Cost of Distributed Learning

Carlee Joe-Wong • Professor, CMU

Thursday, October 1, 2020 at 4:30 PM • Online (Zoom)

View details →

Quantum Renyi relative entropies and their use

Mark Wilde • Professor, Louisiana State University

Thursday, September 17, 2020 at 4:30 PM • Online (Zoom)

View details →

[CANCELLED]

Behnam Neyshabur • Senior Research Scientist, Google

Thursday, June 4, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Moritz Hardt • Professor, UC Berkeley

Thursday, May 28, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Jason Marden • Professor, UC Santa Barbara

Thursday, May 21, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

David Woodruff • Professor, CMU

Thursday, May 14, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Pablo Parillo • Professor, MIT

Thursday, May 7, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Carlee Joe Wong • Professor, CMU

Thursday, April 30, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Baosen Zhang • Professor, University of Washington

Thursday, April 23, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Mark Wilde • Professor, Louisiana State University

Thursday, April 16, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Adam Wierman • Professor, Caltech

Thursday, April 9, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED] Global Optimality Guarantees for Policy Gradient Methods

Daniel Russo • Professor, Columbia

Thursday, March 12, 2020 at 4:30 PM • Packard 101

View details →

Codification Design in Compressive Imaging

Henry Arguello Fuentes • Professor, Universidad Industrial de Santander

Thursday, March 5, 2020 at 4:30 PM • Packard 101

View details →

Empirical Risk Minimization in High-dimensions: Asymptotics, Optimality and Double Descent

Christos Thrampoulidis • Professor, UC Santa Barbara

Thursday, February 27, 2020 at 4:30 PM • Packard 101

View details →

Network Systems, Kuramoto Oscillators, and Synchronous Power Flow

Francesco Bullo • Professor, UC Santa Barbara

Thursday, February 20, 2020 at 4:30 PM • Packard 101

View details →

Recurrent Switching Linear Dynamical Systems for Neural and Behavioral Analysis

Scott Linderman • Professor, Stanford

Thursday, February 13, 2020 at 4:30 PM • Packard 101

View details →

How to trap a gradient flow

Sébastien Bubeck • Senior Researcher, Microsoft Research Redmond

Thursday, February 6, 2020 at 4:30 PM • Packard 101

View details →

Slides

From Feedforward-Designed Convolutional Neural Networks (FF-CNNs) to Successive Subspace Learning (SSL)

C.-C. Jay Kuo • Professor, USC

Thursday, January 30, 2020 at 4:30 PM • Packard 101

View details →

Self-Programming Networks: Applications to Financial Trading Systems

Balaji Prabhakar • Professor, Stanford

Thursday, January 23, 2020 at 4:30 PM • Packard 101

View details →

Theoretical Reflections on Quantum Supremacy

Umesh Vazirani • Professor, UC Berkeley

Thursday, January 16, 2020 at 4:30 PM • Packard 101

View details →

Slides

The Robustness Problem

Justin Gilmer • Research Scientist, Google Brain

Thursday, January 9, 2020 at 4:30 PM • Packard 101

View details →

Past Talks

All past talks organized by academic quarter. For talks prior to 2020, visit our historical archive.

Qualia Optimization: Exploring Mathematical Formulations of AI Experience

Philip Thomas • Associate Professor, University of Massachusetts

Thursday, February 19, 2026 at 4:00 PM • Packard 202

View details →

The Arrow-Hurwicz-Uzawa Method: Historical Perspectives and Current Advances

Xiaoming Yuan • Professor, The University of Hong Kong

Thursday, February 12, 2026 at 4:00 PM • Packard 202

View details →

Controllable AI: Control Theory meets Artificial Intelligence

Carmen Amo Alonso • Postdoc, Stanford University

Thursday, February 5, 2026 at 4:00 PM • Packard 202

View details →

Post-training diffusion models: a unified view from stochastic analysis

Renyuan Xu • Assistant Professor, Stanford University

Thursday, January 22, 2026 at 4:00 PM • Packard 202

View details →

From Bits to Rounds: Unifying Information-Theoretic Limits and Real-World Efficiency in Diffusion Language Models

Jiantao Jiao • Assistant Professor, UC Berkeley

Thursday, November 20, 2025 at 4:00 PM • Packard 202

View details →

PEPFlow: A library for the systematic analysis of optimization algorithms

Xin Jiang • Assistant Professor, University of Houston

Thursday, November 13, 2025 at 4:00 PM • Packard 202

View details →

Subspace injections

Joel Tropp • Professor, Caltech

Thursday, November 6, 2025 at 4:00 PM • Hewlett 201

View details →

Simplifying Complex Machine Learning Models with Mathematical Optimization

Rahul Mazumder • Associate Professor, MIT

Thursday, October 23, 2025 at 4:00 PM • Packard 202

View details →

Navigating Underspecified Optimization Problems using Portfolios for Human-AI Augmented Decision-Making

Swati Gupta • Associate Professor, MIT

Thursday, October 16, 2025 at 4:00 PM • Packard 202

View details →

Multi-period mixed-integer quadratic programming

Alper Atamturk • Professor, UC Berkeley

Thursday, October 9, 2025 at 4:00 PM • Packard 202

View details →

Slides

Randomized Algorithms for Solving Linear Systems with Low-Rank Structure

Michał Dereziński • Assistant Professor, University of Michigan

Thursday, October 2, 2025 at 4:00 PM • Packard 202

View details →

Toward Trustworthy AI: Principled and Automated Interpretability in Deep Learning

Lily Weng • Assistant Professor, UCSD

Thursday, July 31, 2025 at 4:00 PM • Packard 202

View details →

Negative Stepsizes Make Gradient-Descent-Ascent Converge

Jason Altschuler • Assistant Professor, University of Pennsylvania

Thursday, May 29, 2025 at 4:00 PM • Packard 202

View details →

Rank Overparameterization: Embracing Nonconvexity in Low-rank Optimization

Richard Y. Zhang • Assistant Professor, University of Illinois Urbana-Champaign

Thursday, March 6, 2025 at 4:00 PM • Packard 202

View details →

Differentially private M-estimation via noisy optimization

Po-Ling Loh • Professor, University of Cambridge

Thursday, February 27, 2025 at 4:00 PM • Packard 202

View details →

Exponentially faster gradient methods in the presence of ravines

Damek Davis • Visiting Associate Professor, University of Pennsylvania

Thursday, November 21, 2024 at 4:00 PM • Packard 202

View details →

Towards Pluralistic Alignment: Foundations for Learning Diverse Human Preferences

Ramya Korlakai Vinayak • Assistant Professor, University of Wisconsin-Madison

Thursday, November 14, 2024 at 4:00 PM • Packard 202

View details →

Control Across Scales and Applications to Traffic Management

Murat Arcak • Professor, UC Berkeley

Thursday, May 30, 2024 at 4:00 PM • Packard 202

View details →

Cardinal-Utility Matching Markets: From Tractability to Intractability ... and Back!

Vijay Vazirani • Distinguished Professor, UC Irvine

Thursday, May 23, 2024 at 4:00 PM • Packard 101

View details →

Challenges & Opportunities for Learning in Strategic Environments

Eric Mazumdar • Assistant Professor, Caltech

Thursday, April 25, 2024 at 4:00 PM • Packard 202

View details →

What makes learning to control easy or hard?

Nikolai Matni • Assistant Professor, University of Pennsylvania

Thursday, April 18, 2024 at 4:00 PM • Packard 202

View details →

Slides

Optimization, Robustness and Attention in Deep Learning: Insights from Random and NTK Feature Models

Marco Mondelli • Assistant Professor, IST Austria

Thursday, April 11, 2024 at 4:00 PM • Packard 202

View details →

Design and Analysis of Decentralized Learning Enabled Multi-Agent Systems

Chinmay Maheshwari • PhD Student, UC Berkeley

Thursday, March 14, 2024 at 4:00 PM • Packard 202

View details →

What Algorithms can Transformers Learn? A Study in Length Generalization

Preetum Nakkiran • Research Scientist, Apple Research

Thursday, March 7, 2024 at 4:00 PM • Packard 202

View details →

From Large to Small Datasets: Size Generalization for Clustering Algorithm Selection

Ellen Vitercik • Assistant Professor, Stanford

Thursday, February 29, 2024 at 4:00 PM • Packard 202

View details →

Solving semirandom planted CSPs via SDP-certificates and spectral sparsification

Venkatesan Guruswami • Professor, UC Berkeley

Thursday, February 22, 2024 at 4:00 PM • Packard 202

View details →

Clustering for causal inference under network interference

Johan Ugander • Associate Professor, Stanford

Thursday, February 15, 2024 at 4:00 PM • Packard 202

View details →

Permissionless Blockchains: Fundamental Trade-offs in Throughput, Latency, and Safety

Dongning Guo • Professor, Northwestern University

Thursday, February 8, 2024 at 4:00 PM • Packard 202

View details →

Asymptotic Learning in Overparameterized Models

Anant Sahai • Professor, UC Berkeley

Thursday, January 25, 2024 at 4:00 PM • Packard 202

View details →

Model Selection And Ensembling When There Are More Parameters Than Data

Michael Mahoney • Professor, ICSI, LBNL, and UC Berkeley

Thursday, December 7, 2023 at 4:00 PM • Packard 202

View details →

The Economics of Automated Market Making and Decentralized Exchanges

Ciamac Moallemi • Professor, Columbia University

Thursday, November 30, 2023 at 4:00 PM • Packard 202

View details →

Efficient Neural Scene Representation, Rendering, and Generation

Gordon Wetzstein • Associate Professor, Stanford

Thursday, November 16, 2023 at 4:00 PM • Packard 202

View details →

Compressed Sensing using Generative Models: Theory and Applications

Ajil Jalal • Postdoc, UC Berkeley

Thursday, November 9, 2023 at 4:00 PM • Packard 202

View details →

See versus Hear - Wireless Digital Traffic Signs

Chai Toh • Senior Fellow, UC Berkeley

Thursday, November 2, 2023 at 4:00 PM • Packard 202

View details →

From Optimization to Sampling, with a touch of Schrödinger Bridge

Qijia Jiang • Postdoc, UC Berkeley

Thursday, October 26, 2023 at 4:00 PM • Packard 202

View details →

CICERO: Human-Level Performance in the Game of Diplomacy by Combining Language Models with Strategic Reasoning

Noam Brown • Research Scientist, OpenAI

Thursday, October 19, 2023 at 4:00 PM • Packard 202

View details →

Toward a grand unified theory of accelerations in optimization and machine learning

Ernest Ryu • Assistant Professor, Seoul National University

Thursday, October 12, 2023 at 4:00 PM • Packard 202

View details →

Low rank approximation for faster optimization

Madeleine Udell • Assistant Professor, Stanford

Thursday, June 1, 2023 at 4:00 PM • Packard 202

View details →

DoG is SGD's best friend: toward tuning-free stochastic optimization

Yair Carmon • Assistant Professor, Tel Aviv University

Wednesday, May 31, 2023 at 2:00 PM • Packard 202

View details →

Mathematical underpinnings of emerging computational imaging inverse problems

Shirin Jalali • Assistant Professor, Rutgers University

Thursday, May 25, 2023 at 4:00 PM • Packard 202

View details →

Automating the analysis and design of iterative optimization algorithms

Laurent Lessard • Associate Professor, Northeastern University

Thursday, May 18, 2023 at 4:00 PM • Packard 202

View details →

Representative and Deliberative Social Choice

Ashish Goel • Professor, Stanford

Thursday, April 27, 2023 at 4:00 PM • Packard 202

View details →

New local differentially private protocols for frequency and mean estimation

Jelani Nelson • Professor, UC Berkeley

Friday, April 21, 2023 at 4:00 PM • Packard 202

View details →

Near-Optimal Algorithms for Imitation Learning

Jiantao Jiao • Assistant Professor, UC Berkeley

Thursday, April 13, 2023 at 2:00 PM • Packard 202

View details →

Breaking the Sample Size Barrier in Reinforcement Learning

Yuxin Chen • Associate Professor, University of Pennsylvania

Thursday, April 6, 2023 at 4:00 PM • Packard 202

View details →

Looking beyond the Worst-Case Adversaries in Machine Learning

Nika Haghtalab • Assistant Professor, UC Berkeley

Thursday, March 23, 2023 at 4:00 PM • Packard 202

View details →

An Information Theory for Out-of-Order Information: Applications in DNA Data Storage and Beyond

Ilan Shomorony • Assistant Professor, UIUC

Thursday, March 16, 2023 at 4:00 PM • Packard 202

View details →

Kernel Thinning and Stein Thinning

Lester Mackey • Principal Researcher, Microsoft Research

Thursday, March 9, 2023 at 4:00 PM • Packard 202

View details →

Slides

Causal Representation Learning – A Proposal

Caroline Uhler • Professor, MIT

Friday, March 3, 2023 at 1:00 PM • Packard 202

View details →

Gromov-Wasserstein Distances: Entropic Regularization, Duality, and Sample Complexity

Ziv Goldfeld • Assistant Professor, Cornell University

Thursday, February 23, 2023 at 4:00 PM • Packard 202

View details →

Adversarial machine learning and instrumental variables for flexible causal modeling

Vasilis Syrgkanis • Assistant Professor, Stanford

Thursday, February 2, 2023 at 4:00 PM • Packard 202

View details →

Learning with Feature Geometry

Lizhong Zheng • Professor, MIT

Thursday, January 26, 2023 at 4:00 PM • Packard 202

View details →

Recent Advances in Non-log-concave and Heavy-tailed Sampling

Krishna Balasubramanian • Assistant Professor, UC Davis

Thursday, January 19, 2023 at 4:00 PM • Packard 202

View details →

Next-Generation Communication/Artificial-Intelligence Challenges

John Cioffi • Professor, Stanford

Thursday, January 12, 2023 at 4:00 PM • Packard 202

View details →

What kind of information is there in single-cell genomics data?

Lior Pachter • Professor, Caltech

Thursday, December 8, 2022 at 4:00 PM • Packard 202

View details →

Fundamental limits of low-rank matrix estimation with diverging aspect ratios

Yuchen Wu • PhD student, Stanford

Thursday, December 1, 2022 at 4:00 PM • Packard 202

View details →

Two Vignettes on Efficient Procedures for Personalized Decision Making

Raaz Dwivedi • Postdoc, Harvard/MIT

Thursday, November 17, 2022 at 4:00 PM • Packard 202

View details →

Recent Advances in Colonel Blotto Games and the Connection to Controls

Jason R. Marden • Professor, UCSB

Thursday, November 10, 2022 at 4:00 PM • Packard 202

View details →

Leveraging 'partial' smoothness for faster convergence in nonsmooth optimization

Damek Davis • Associate Professor, Cornell

Thursday, November 3, 2022 at 4:00 PM • Packard 202

View details →

Towards Instance-Optimal Algorithms for Reinforcement Learning

Kevin Jamieson • Assistant Professor, University of Washington

Thursday, October 27, 2022 at 4:00 PM • Packard 202

View details →

A data-centric view on reliable generalization

Ludwig Schmidt • Assistant Professor, University of Washington

Thursday, October 20, 2022 at 4:00 PM • Packard 202

View details →

Stochastic Algorithms in the Large: Batch Size Saturation, Stepsize Criticality, Generalization Performance, and Exact Dynamics

Courtney Paquette • Assistant Professor, McGill University

Thursday, October 13, 2022 at 4:00 PM • Packard 101

View details →

Computational aspects of learning sparse functions with neural networks

Theodor Misiakiewicz • PhD student, Stanford

Thursday, October 6, 2022 at 4:00 PM • Packard 101

View details →

Attacking the privacy of machine learning models

Nicholas Carlini • Research Scientist, Google Brain

Thursday, September 29, 2022 at 4:00 PM • Packard 101

View details →

Algorithms and computational limits for infinite-horizon general-sum stochastic games

Vidya Muthukumar • Professor, Georgia Tech

Thursday, May 19, 2022 at 4:00 PM • Packard 101

View details →

Beyond the Csiszár-Körner Bound: Best-Possible Wiretap Coding via Obfuscation

Amit Sahai • Professor, UCLA

Thursday, May 12, 2022 at 4:00 PM • Packard 101

View details →

The online convex optimization approach to control

Elad Hazan • Professor, Princeton

Thursday, May 5, 2022 at 4:00 PM • Packard 101

View details →

Two vignettes about interpolation and generalization in overparameterized models

Niladri Chatterji • Postdoc, Stanford

Thursday, April 28, 2022 at 4:00 PM • Packard 101

View details →

Historical thoughts on modern prediction

Ben Recht • Professor, UC Berkeley

Thursday, April 21, 2022 at 4:00 PM • Packard 101

View details →

What is the Statistical Complexity of Reinforcement Learning?

Sham Kakade • Professor, Harvard

Thursday, April 14, 2022 at 1:00 PM • Online (Zoom)

View details →

Palo Alto, We Have a Problem. There Is No Oracle!

Amin Karbasi • Professor, Yale University

Thursday, March 17, 2022 at 4:00 PM • Online (Zoom)

View details →

Towards instance-optimal compression for distributed mean estimation

Ananda Theertha Suresh • Research Scientist, Google Research

Thursday, March 10, 2022 at 4:00 PM • Online (Zoom)

View details →

Adaptivity and Confounding in Multi-Armed Bandit Experiments

Daniel Russo • Professor, Columbia Business School

Thursday, March 3, 2022 at 4:00 PM • Online (Zoom)

View details →

Statistical approaches for mechanistic biological discovery with genomics

Julia Salzman • Professor, Stanford

Thursday, February 24, 2022 at 4:00 PM • Online (Zoom)

View details →

Denoising as a Building Block for Imaging, Inverse Problems, and Machine Learning

Peyman Milanfar • Principal Scientist / Director, Google Research

Thursday, February 10, 2022 at 4:00 PM • Online (Zoom)

View details →

A Continuized View on Nesterov Acceleration for Stochastic Gradient Descent and Randomized Gossip

Raphaël Berthier • Postdoc, EPFL

Thursday, February 3, 2022 at 4:00 PM • Packard 101

View details →

The Foundations of Private Statistical Estimation

Jonathan Ullman • Professor, Northeastern University

Thursday, January 13, 2022 at 4:00 PM • Packard 101

View details →

Contextual Search in the Presence of Irrational Agents

Chara Podimata • Graduate Student, Harvard

Thursday, January 6, 2022 at 4:00 PM • Packard 101

View details →

Connecting Statistical Problems with Different Structures

Guy Bresler • Professor, MIT

Thursday, December 2, 2021 at 4:00 PM • Packard 101

View details →

Optimization in Theory and Practice

Stephen Wright • Professor, University of Wisconsin-Madison

Thursday, November 18, 2021 at 4:00 PM • Packard 101

View details →

Safe and Efficient Reinforcement Learning for Power System Control

Baosen Zhang • Professor, University of Washington

Thursday, November 11, 2021 at 4:00 PM • Packard 101

View details →

Distributed Algorithms for Optimization in Networks

Angelia Nedich • Professor, Arizona State University

Thursday, November 4, 2021 at 4:00 PM • Packard 101

View details →

On the Convergence of Langevin Monte Carlo: The Interplay between Tail Growth and Smoothness

Murat Erdogdu • Assistant Professor, University of Toronto

Thursday, October 28, 2021 at 4:00 PM • Packard 101

View details →

AI for clinical trials and clinical trials for AI

James Zou • Professor, Stanford

Thursday, October 21, 2021 at 4:00 PM • Packard 101

View details →

Experimentation and Decision-Making in Two-Sided Marketplaces: The Impact of Interference / Simple Agent, Complex Environment: Efficient Reinforcement Learning with Agent States

Hannah Li / Shi Dong • PhD Students, Stanford

Thursday, October 14, 2021 at 4:00 PM • Packard 101

View details →

KO codes: Inventing Non-linear Encoding and Decoding for Reliable Wireless Communication via Deep-learning

Ashok Vardhan Makkuva • PhD Student, UIUC

Thursday, October 7, 2021 at 4:00 PM • Packard 101

View details →

Sequential Decision Making: How Much Adaptivity Is Needed Anyways?

Amin Karbasi • Professor, Yale

Thursday, September 30, 2021 at 4:45 PM • Packard 101

View details →

Interpolation Phase Transition in Neural Networks: Memorization and Generalization

Joe Zhong • Postdoc, Stanford

Thursday, September 23, 2021 at 4:30 PM • Packard 101

View details →

Between tractable and intractable problems in reinforcement learning

Csaba Szepesvari • Professor, University of Alberta

Thursday, June 10, 2021 at 4:30 PM • Online (Zoom)

View details →

The Hidden Convex Optimization Landscape of Deep Neural Networks

Mert Pilanci • Professor, Stanford

Thursday, June 3, 2021 at 4:30 PM • Online (Zoom)

View details →

Learning from preferences and labels

Aarti Singh • Professor, Carnegie Mellon University

Thursday, May 27, 2021 at 4:30 PM • Online (Zoom)

View details →

Multi-Scale Methods for Machine Learning

Inderjit S. Dhillon • Professor, UT Austin

Thursday, May 20, 2021 at 4:30 PM • Online (Zoom)

View details →

Interior Point Methods for Nearly Linear Time Algorithms

Aaron Sidford • Professor, Stanford

Thursday, May 6, 2021 at 4:30 PM • Online (Zoom)

View details →

Private Stochastic Convex Optimization

Kunal Talwar • Research Scientist, Apple

Thursday, April 29, 2021 at 4:30 PM • Online (Zoom)

View details →

Recent results in planted assignment problems

Yihong Wu • Professor, Yale

Thursday, April 22, 2021 at 4:30 PM • Online (Zoom)

View details →

Demystfying the Efficiency of Reinforcement Learning: Two Recent Stories

Yuxin Chen • Professor, Princeton

Thursday, April 15, 2021 at 4:30 PM • Online (Zoom)

View details →

Scalable semidefinite programming

Joel Tropp • Professor, Caltech

Thursday, April 8, 2021 at 4:30 PM • Online (Zoom)

View details →

Finding Global Minima via Kernel Approximations

Francis Bach • Researcher, INRIA

Thursday, April 1, 2021 at 10:00 AM • Online (Zoom)

View details →

Approximating cross-validation: guarantees for model assessment and selection

Ashia Wilson • Professor, MIT

Thursday, March 18, 2021 at 4:30 PM • Online (Zoom)

View details →

Robust Learning from Batches -- The Best Things in Life are (Almost) Free

Alon Orlitsky • Professor, UC San Diego

Thursday, March 11, 2021 at 4:30 PM • Online (Zoom)

View details →

Exploiting Symmetries in Inference and Learning

Max Welling • Professor, University of Amsterdam

Thursday, March 4, 2021 at 9:00 AM • Online (Zoom)

View details →

Chasing the Long Tail: What Neural Networks Memorize and Why

Vitaly Feldman • Research Scientist, Apple AI Research

Thursday, February 25, 2021 at 4:30 PM • Online (Zoom)

View details →

Slides

Enabling Fast and Robust Federated Learning

Ramtin Pedarsani • Professor, UC Santa Barbara

Thursday, February 18, 2021 at 4:30 PM • Online (Zoom)

View details →

Data, decisions, and dynamics

Moritz Hardt • Professor, UC Berkeley

Thursday, February 11, 2021 at 4:30 PM • Online (Zoom)

View details →

End-to-end learning for computational microscopy

Laura Waller • Professor, UC Berkeley

Thursday, February 4, 2021 at 4:30 PM • Online (Zoom)

View details →

The Well Tempered Lasso

Yoram Singer • Professor, Princeton University

Thursday, January 28, 2021 at 4:30 PM • Online (Zoom)

View details →

Adaptive Experimental Design for Best Identification and Multiple Testing

Kevin Jamieson • Professor, University of Washington

Thursday, January 21, 2021 at 4:30 PM • Online (Zoom)

View details →

Accelerating Reinforcement Learning in Emerging Wireless IoT Systems and Applications via System Awareness

Nicholas Mastronarde & Jacob Chakareski • Professors, State University of New York at Buffalo & New Jersey Institute of Technology

Thursday, January 14, 2021 at 4:30 PM • Online (Zoom)

View details →

Interference in Experimental Design in Online Platforms

Ramesh Johari • Professor, Stanford

Thursday, November 19, 2020 at 4:30 PM • Online (Zoom)

View details →

A Very Sketchy Talk

David Woodruff • Professor, CMU

Friday, November 13, 2020 at 1:15 PM • Online (Zoom)

View details →

Learning Convolutions from Scratch

Behnam Neyshabur • Sr. Research Scientist, Google

Thursday, November 5, 2020 at 4:30 PM • Online (Zoom)

View details →

Towards Model Agnostic Robustness

Sujay Sanghavi • Professor, UT Austin

Thursday, October 29, 2020 at 4:30 PM • Online (Zoom)

View details →

Competitive Control via Online Optimization

Adam Wierman • Professor, Caltech

Thursday, October 22, 2020 at 4:30 PM • Online (Zoom)

View details →

Strategies for Active Machine Learning

Robert Nowak • Professor, University of Wisconsin-Madison

Thursday, October 15, 2020 at 4:30 PM • Online (Zoom)

View details →

Computational Barriers to Estimation from Low-Degree Polynomials

Alex Wein • Instructor, NYU Courant Institute

Thursday, October 8, 2020 at 4:30 PM • Online (Zoom)

View details →

Optimizing the Cost of Distributed Learning

Carlee Joe-Wong • Professor, CMU

Thursday, October 1, 2020 at 4:30 PM • Online (Zoom)

View details →

Quantum Renyi relative entropies and their use

Mark Wilde • Professor, Louisiana State University

Thursday, September 17, 2020 at 4:30 PM • Online (Zoom)

View details →

[CANCELLED]

Behnam Neyshabur • Senior Research Scientist, Google

Thursday, June 4, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Moritz Hardt • Professor, UC Berkeley

Thursday, May 28, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Jason Marden • Professor, UC Santa Barbara

Thursday, May 21, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

David Woodruff • Professor, CMU

Thursday, May 14, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Pablo Parillo • Professor, MIT

Thursday, May 7, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Carlee Joe Wong • Professor, CMU

Thursday, April 30, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Baosen Zhang • Professor, University of Washington

Thursday, April 23, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Mark Wilde • Professor, Louisiana State University

Thursday, April 16, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED]

Adam Wierman • Professor, Caltech

Thursday, April 9, 2020 at 4:30 PM • Packard 101

View details →

[CANCELLED] Global Optimality Guarantees for Policy Gradient Methods

Daniel Russo • Professor, Columbia

Thursday, March 12, 2020 at 4:30 PM • Packard 101

View details →

Codification Design in Compressive Imaging

Henry Arguello Fuentes • Professor, Universidad Industrial de Santander

Thursday, March 5, 2020 at 4:30 PM • Packard 101

View details →

Empirical Risk Minimization in High-dimensions: Asymptotics, Optimality and Double Descent

Christos Thrampoulidis • Professor, UC Santa Barbara

Thursday, February 27, 2020 at 4:30 PM • Packard 101

View details →

Network Systems, Kuramoto Oscillators, and Synchronous Power Flow

Francesco Bullo • Professor, UC Santa Barbara

Thursday, February 20, 2020 at 4:30 PM • Packard 101

View details →

Recurrent Switching Linear Dynamical Systems for Neural and Behavioral Analysis

Scott Linderman • Professor, Stanford

Thursday, February 13, 2020 at 4:30 PM • Packard 101

View details →

How to trap a gradient flow

Sébastien Bubeck • Senior Researcher, Microsoft Research Redmond

Thursday, February 6, 2020 at 4:30 PM • Packard 101

View details →

Slides

From Feedforward-Designed Convolutional Neural Networks (FF-CNNs) to Successive Subspace Learning (SSL)

C.-C. Jay Kuo • Professor, USC

Thursday, January 30, 2020 at 4:30 PM • Packard 101

View details →

Self-Programming Networks: Applications to Financial Trading Systems

Balaji Prabhakar • Professor, Stanford

Thursday, January 23, 2020 at 4:30 PM • Packard 101

View details →

Theoretical Reflections on Quantum Supremacy

Umesh Vazirani • Professor, UC Berkeley

Thursday, January 16, 2020 at 4:30 PM • Packard 101

View details →

Slides

The Robustness Problem

Justin Gilmer • Research Scientist, Google Brain

Thursday, January 9, 2020 at 4:30 PM • Packard 101

View details →

Upcoming Talks

Winter 2026

Qualia Optimization: Exploring Mathematical Formulations of AI Experience

The Arrow-Hurwicz-Uzawa Method: Historical Perspectives and Current Advances

Controllable AI: Control Theory meets Artificial Intelligence

Post-training diffusion models: a unified view from stochastic analysis

Fall 2025

From Bits to Rounds: Unifying Information-Theoretic Limits and Real-World Efficiency in Diffusion Language Models

PEPFlow: A library for the systematic analysis of optimization algorithms

Subspace injections

Simplifying Complex Machine Learning Models with Mathematical Optimization

Navigating Underspecified Optimization Problems using Portfolios for Human-AI Augmented Decision-Making

Multi-period mixed-integer quadratic programming

Randomized Algorithms for Solving Linear Systems with Low-Rank Structure

Summer 2025

Toward Trustworthy AI: Principled and Automated Interpretability in Deep Learning

Spring 2025

Negative Stepsizes Make Gradient-Descent-Ascent Converge

Winter 2025

Rank Overparameterization: Embracing Nonconvexity in Low-rank Optimization

Differentially private M-estimation via noisy optimization

Fall 2024

Exponentially faster gradient methods in the presence of ravines

Towards Pluralistic Alignment: Foundations for Learning Diverse Human Preferences

Spring 2024

Control Across Scales and Applications to Traffic Management

Cardinal-Utility Matching Markets: From Tractability to Intractability ... and Back!

Challenges & Opportunities for Learning in Strategic Environments

What makes learning to control easy or hard?

Optimization, Robustness and Attention in Deep Learning: Insights from Random and NTK Feature Models

Winter 2024

Design and Analysis of Decentralized Learning Enabled Multi-Agent Systems

What Algorithms can Transformers Learn? A Study in Length Generalization

From Large to Small Datasets: Size Generalization for Clustering Algorithm Selection

Solving semirandom planted CSPs via SDP-certificates and spectral sparsification

Clustering for causal inference under network interference

Permissionless Blockchains: Fundamental Trade-offs in Throughput, Latency, and Safety

Asymptotic Learning in Overparameterized Models

Fall 2023

Model Selection And Ensembling When There Are More Parameters Than Data

The Economics of Automated Market Making and Decentralized Exchanges

Efficient Neural Scene Representation, Rendering, and Generation

Compressed Sensing using Generative Models: Theory and Applications

See versus Hear - Wireless Digital Traffic Signs

From Optimization to Sampling, with a touch of Schrödinger Bridge

CICERO: Human-Level Performance in the Game of Diplomacy by Combining Language Models with Strategic Reasoning

Toward a grand unified theory of accelerations in optimization and machine learning

Spring 2023

Low rank approximation for faster optimization

DoG is SGD's best friend: toward tuning-free stochastic optimization

Mathematical underpinnings of emerging computational imaging inverse problems

Automating the analysis and design of iterative optimization algorithms

Representative and Deliberative Social Choice

New local differentially private protocols for frequency and mean estimation

Near-Optimal Algorithms for Imitation Learning

Breaking the Sample Size Barrier in Reinforcement Learning

Winter 2023

Looking beyond the Worst-Case Adversaries in Machine Learning

An Information Theory for Out-of-Order Information: Applications in DNA Data Storage and Beyond

Kernel Thinning and Stein Thinning

Causal Representation Learning – A Proposal

Gromov-Wasserstein Distances: Entropic Regularization, Duality, and Sample Complexity

Adversarial machine learning and instrumental variables for flexible causal modeling

Learning with Feature Geometry

Recent Advances in Non-log-concave and Heavy-tailed Sampling

Next-Generation Communication/Artificial-Intelligence Challenges

Fall 2022

What kind of information is there in single-cell genomics data?

Fundamental limits of low-rank matrix estimation with diverging aspect ratios

Two Vignettes on Efficient Procedures for Personalized Decision Making

Recent Advances in Colonel Blotto Games and the Connection to Controls

Leveraging 'partial' smoothness for faster convergence in nonsmooth optimization

Towards Instance-Optimal Algorithms for Reinforcement Learning

A data-centric view on reliable generalization

Stochastic Algorithms in the Large: Batch Size Saturation, Stepsize Criticality, Generalization Performance, and Exact Dynamics

Computational aspects of learning sparse functions with neural networks

Attacking the privacy of machine learning models

Spring 2022

Algorithms and computational limits for infinite-horizon general-sum stochastic games

Beyond the Csiszár-Körner Bound: Best-Possible Wiretap Coding via Obfuscation