Conference Schedule

Times are in Central European Time; All sessions are in the same room “Salle Dussane”. Breaks are in the “Rotonde” (the small room behind).

Note that France has its daylight saving time change on Sunday, March 27 (right before the conference). For instance, the time difference between Paris and New-York will be of 6 hours during the conference (and not just 5).

The link to access the conference is the following one It will ask for your registration e-mail, and send you a confirmation code that will let you in the virtual space.


Tuesday, March 29, 2022

13:50-14:00Welcome notes, (Nika Haghtalab, Sanjoy Dasgupta)
14:00-15:00Session 1, in person (Chair: Satyen Kale)
Scale-Free Adversarial Multi Armed Bandits
Sudeep Raja Putta (Columbia University); Shipra Agrawal (Columbia University)
Distributed Online Learning for Joint Regret with Communication Constraints
Dirk van der Hoeven (Università degli Studi di Milano); Hedi Hadiji (University of Amsterdam); Tim van Erven (University of Amsterdam)
Social Learning in Non-Stationary Environments
Etienne Boursier (EPFL); Vianney Perchet (ENSAE & Criteo AI Lab); Marco Scarsini (LUISS)
15:30-16:30Session 2, virtual (Chair: Quentin Berthet)
Decentralized Cooperative Reinforcement Learning with Hierarchical Information Structure
Hsu Kao (University of Michigan); Chen-Yu Wei (University of Southern California); Vijay Subramanian (University of Michigan)
Algorithms for learning a mixture of linear classifiers
Aidao Chen (Northwestern University); Anindya De (University of Pennsylvania); Aravindan Vijayaraghavan (Northwestern University)
Learning with distributional inverters
Eric R Binnendyk (New Mexico Institute of Mining and Technology); Marco L Carmosino (Boston University); Antonina Kolokolova (Memorial University Newfoundland); Ramyaa Ramyaa (New Mexico Institute of Mining and Technology); Manuel Sabin (UC Berkeley)
17:00-18:00Plenary talk, virtual (Chair: Nika Haghtalab)

Sebastien Bubeck (MSR)
Chasing sets, with an application to online shortest path
18:30-21:00Welcome Reception (la Rotonde)

Wednesday, March 30, 2022

9:30-10:30Session 1, in person (Chair: Tim van Erven)
Global Riemannian Acceleration in Hyperbolic and Spherical Spaces
David Martínez-Rubio (University of Oxford)
Iterated Vector Fields and Conservatism, with Applications to Federated Learning
Zachary Charles (Google Research); John K Rush (Google Research)
Inductive Bias of Gradient Descent for Weight Normalized Smooth Homogeneous Neural Nets
Depen Morwani (Indian Institute of Technology, Madras); Harish Guruprasad Ramaswamy (IIT Madras)
11:00-12:00Session 2, virtual (Chair: Sanjoy Dasgupta)
Efficient Local Planning with Linear Function Approximation
Dong Yin (DeepMind); Botao Hao (Deepmind); Yasin Abbasi-Yadkori (DeepMind); Nevena Lazic (DeepMind); Csaba Szepesvari (DeepMind/University of Alberta)
Faster Noisy Power Method
Zhiqiang Xu (Baidu); Ping Li (Baidu Research)
Refined Lower Bounds for Nearest Neighbor Condensation
Rajesh Chitnis (University of Birmingham)
14:00-15:00Business Meeting, hybrid (Chair: Mehryar Mohri)
15:30-16:30Session 4, virtual (Chair: Vianney Perchet)
Polynomial-Time Sum-of-Squares Can Robustly Estimate Mean and Covariance of Gaussians Optimally
Peter Manohar (Carnegie Mellon University); Pravesh Kothari (Carnegie Mellon University); Brian H Zhang (Carnegie Mellon University)
Understanding Simultaneous Train and Test Robustness
Pranjal Awasthi (Google); Sivaraman Balakrishnan (Carnegie Mellon University); Aravindan Vijayaraghavan (Northwestern University)
Improved rates for prediction and identification of partially observed linear dynamical systems
Holden Lee (Duke)
17:00-18:00Session 5, virtual (Chair: Nika Haghtalab)
Faster Perturbed Stochastic Gradient Methods for Finding Local Minima
Zixiang Chen (UCLA); Dongruo Zhou (UCLA); Quanquan Gu (University of California, Los Angeles)
Distinguishing Relational Pattern Languages With a Small Number of Short Strings
Robert Holte (University of Alberta); Seyyedmahmoud Mousavi (University of Regina); Sandra Zilles (University of Regina, Canada)
Lower Bounds on the Total Variation Distance Between Mixtures of Two Gaussians
Sami Davies (UW); Arya Mazumdar (University of California, San Diego); Cyrus Rashtchian (Google Research); Soumyabrata Pal (University of Massachusetts Amherst)

Thursday, March 31, 2022

9:30-10:30Session 1, in person (Chair: Claire Vernade)
Metric Entropy Duality and the Sample Complexity of Outcome Indistinguishability
Lunjia Hu (Stanford University); Charlotte Peale (Stanford University); Omer Reingold (Stanford University)
TensorPlan and the Few Actions Lower Bound for Planning in MDPs under Linear Realizability of Optimal Value Functions
Gellert Weisz (DeepMind, UCL); Csaba Szepesvari (DeepMind/University of Alberta); Andras Gyorgy (DeepMind)
Asymptotic Degradation of Linear Regression Estimates with Strategic Data Sources
Benjamin Roussillon (Université Grenoble-Alpes); Nicolas Gast (INRIA); Patrick Loiseau (Inria); Panayotis Mertikopoulos (CNRS and Criteo AI Lab)
11:00-12:00Session 2, in person (Chair: Lev Reyzin)
Efficient Methods for Online Multiclass Logistic Regression
Naman Agarwal (Google); Satyen Kale (Google); Julian Zimmert (Google)
Learning what to remember
Robi Bhattacharjee (University of California, San Diego); Gaurav Mahajan (University of California, San Diego)
Multicalibrated Partitions for Importance Weights
Parikshit Gopalan (VMware Research); Omer Reingold (Stanford University); Vatsal Sharan (USC); Udi Wieder (VMware Research)
14:00-15:00Plenary talk, virtual (Chair: Sanjoy Das Gupta)

Lenka Zdeborova (EPFL)
Understanding trajectories of gradient-based algorithms in solvable non-convex models
15:30-16:30Session 4, virtual (Chair: Sanjoy Das Gupta)
On the Last Iterate Convergence of Momentum Methods
Xiaoyu Li (Boston University); Mingrui Liu (George Mason University); Francesco Orabona (Boston University)
Limiting Behaviors of Nonconvex-Nonconcave Minimax Optimization via Continuous-Time Systems
Benjamin Grimmer (Cornell University); Haihao Lu (University of Chicago); Pratik Worah (Google); Vahab Mirrokni (Google)
On the Initialization for Convex-Concave Min-max Problems
Mingrui Liu (George Mason University); Francesco Orabona (Boston University)
17:00-18:00Session 5, virtual (Chair: Vianney Perchet)
Efficient and Optimal Fixed-Time Regret with Two Experts
Laura Greenstreet (University of British Columbia); Nick Harvey (University of British Columbia); Victor Portella (University of British Columbia)
A Model Selection Approach for Corruption Robust Reinforcement Learning
Chen-Yu Wei (University of Southern California); Chris Dann (Google); Julian Zimmert (Google)
Almost Optimal Algorithms for Two-player Zero-Sum Linear Mixture Markov Games
Zixiang Chen (UCLA); Dongruo Zhou (UCLA); Quanquan Gu (University of California, Los Angeles)
18:30-19:30Best Papers Awards Reception (la Rotonde)
20:00-23:00Social Dinner (Marty)

Friday, April 1, 2022

09:30-10:30Session 1, virtual (Chair: Quentin Berthet)
Minimization by Incremental Stochastic Surrogate Optimization for Large Scale Nonconvex Problems
Belhal Karimi (Baidu Research); Hoi-To Wai (Chinese University of Hong Kong); Eric Moulines (Ecole Polytechnique); Ping Li (Baidu Research)
Faster Rates of Differentially Private Stochastic Convex Optimization
Jinyan Su (University of Electronic Science and Technology of China); Lijie Hu (King Abdullah University of Science and Technology); Di Wang (KAUST)
Adversarial Interpretation of Bayesian Inference
Hisham Husain (Amazon); Jeremias Knoblauch (University College London)
11:00-1200Session 2, in person (Chair: Naman Agarwal)
Infinitely Divisible Noise in the Low Privacy Regime
Nina Mesing Stausholm (IT University of Copenhagen); Rasmus Pagh (University of Copenhagen)
Privacy Amplification via Shuffling for Linear Contextual Bandits
Evrard Garcelon (Facebook); Kamalika Chaudhuri (University of California, San Diego); Vianney Perchet (ENSAE & Criteo AI Lab); Matteo Pirotta (Facebook AI Research)
Beyond Bernoulli: Generating Random Outcomes that cannot be Distinguished from Nature
Cynthia Dwork (Harvard); Michael P Kim (UC Berkeley); Omer Reingold (Stanford University); Guy N Rothblum (Weizmann Institute of Science); Gal Yona (Weizmann Institute of Science)
14:00-15:00Session 3, virtual (Chair: Nika Haghtalab)
Implicit Parameter-free Online Learning with Truncated Linear Models
Keyi Chen (Boston University); Ashok Cutkosky (Boston University); Francesco Orabona (Boston University)
Efficient and Optimal Algorithms for Contextual Dueling Bandits under Realizability
Aadirupa Saha (Microsoft Research); Akshay Krishnamurthy (Microsoft)
The Mirror Langevin Algorithm Converges with Vanishing Bias
Ruilin Li (Georgia Institute of Technology); Molei Tao (Georgia Institute of Technology); Santosh Vempala (Georgia Tech); Andre Wibisono (Yale University)
15:30-16:30Session 4, in person (Chair: Gergo Neu)
Universal Online Learning with Unbounded Losses: Memory Is All You Need
Moise Blanchard (Massachusetts Institute of Technology); Romain Cosson (MIT); Steve Hanneke (Purdue University)
Leveraging Initial Hints for Free in Stochastic Linear Bandits
Richard Zhang (Google Brain); Abhimanyu Das (Google); Ashok Cutkosky (Boston University); Chris Dann (Google)
Universally Consistent Online Learning with Arbitrarily Dependent Responses
Steve Hanneke (Purdue University)