• Click on the titles of individual talks for abstract, slides and archived video.

–
Coffee and Check-In
–
Opening Remarks
–
Data Driven Optimization Models and Algorithms
Yinyu Ye, Stanford University
–
Break
–
Regularized Nonlinear Acceleration
Alex d'Aspremont, École Normale Supérieure
–
Learning One-hidden-layer Neural Networks with Landscape Design
Tengyu Ma, Stanford University
–
Lunch
–
Algorithmic Stability for Interactive Data Analysis
Jonathan Ullman, Northeastern University
–
Preventing Overfitting in Adaptive Data Analysis via Stability
Thomas Steinke, IBM Almaden
–
Break
–
The Convergence of Hamiltonian Monte Carlo
Santosh Vempala, Georgia Institute of Technology
–
Distributional Robustness, Learning, and Empirical Likelihood
John Duchi, Stanford University
–
Reception
–
Coffee and Check-In
–
An Instability in Variational Methods for Learning Topic Models
Andrea Montanari, Stanford University
–
Break
–
A Few Connections Between Optimization and Probability
Sasha Rakhlin, University of Pennsylvania
–
Implicit Regularization in Nonconvex Statistical Estimation
Yuxin Chen, Princeton University
–
Lunch
–
Low-rank Matrix Completion: Adaptive Sampling Can Help When, How?
Rachel Ward, University of Texas at Austin
–
Break
–
Open Problem Session
–
Coffee and Check-In
–
Graph Powering
Emmanuel Abbe, Princeton University
–
Break
–
On Approximation Guarantees for Greedy Low Rank Optimization
Sahand Negahban, Yale University
–
Fast Rates for Bandit Optimization with Upper-Confidence Frank-Wolfe
Quentin Berthet, University of Cambridge
–
Lunch
–
Optimization's Untold Gift to Learning: Implicit Regularization
Nathan Srebro Bartom, Toyota Technological Institute at Chicago
–
Break
–
The Power and Limitations of Kernel Learning
Misha Belkin, Ohio State University
–
Sampling Algorithms for Combinatorial Pure Exploration
Anupam Gupta, Carnegie Mellon University
–
Coffee and Check-In
–
Compositional Properties of Statistical Procedures: An Information-Theoretic View
Max Raginsky, University of Illinois at Urbana-Champaign
–
Break
–
Stability and Convergence Trade-Off of Iterative Optimization Algorithms
Bin Yu, UC Berkeley
–
Accuracy First: Selecting a Differential Privacy Level for Accuracy-Constrained Empirical Risk Minimization
Katrina Ligett, Hebrew University of Jerusalem