Bridging Mathematical Optimization, Information Theory, and Data Science - Schedule

Monday, May 14, 2018 
8:30-9:00 - Check in and Continental Breakfast 
9:00-9:05 - Welcome and introduction 

9:05am - 9:50am

  • Peter Bartlett 
  • Optimization and Generalization Properties of Deep Neural Networks 

9:50am - 10:35am

  • Stephen Wright 
  • Nonconvex Optimization Algorithms with Complexity Guarantees 
10:35-11:00 - Break 

11:00am - 11:30am

  • Yihong Wu 
  • Recovering a Hidden Hamiltonian Cycle via Linear Programming 

11:30am - 12:00pm

  • Alex Dimakis 
  • Gans for Compressed Sensing and Adversarial Defense 

12:00 - 12:30pm

  • Mengdi Wang
  • On the (Reduced) Complexity of Markov Decision Process
12:30-2:00 - LUNCH at PCTS 

2:00pm - 2:30pm

  • Ming Yuan 
  • Statistically and Computationally Efficient Tensor Completion 

2:30pm - 3:15

  • Andrea Montanari 
  • A Mean Field View of the Landscape of Two-Layers Neural Networks 

3:15 - 3:45pm

  • Zongming Ma 
  • Optimal Hypothesis Testing For Stochastic Block Models with Growing Degrees 
3:45-4:15 - Break 

4:15pm - 5:00pm

  • Harrison Zhou 
  • Statistical and Computational Guarantees of Mean Field Variational Inference for Community Detection 

5:00pm - 5:30pm

  • Anru Zhang 
  • Sparse and Low-Rank Tensor Estimation via Cubic Sketchings 

5:30pm - 6:15pm

  • Ankur Moitra 
  • Robustness Meets Algorithms 

Wednesday, May 16, 2018 

8:30-9:00 Continental Breakfast 

8:45am - 9:30am

  • Pramod Vishwanath 
  • Learning in Gated Neural Networks 

9:30am - 10:15am

  • Alex Shapiro 
  • Matrix Completion with Deterministic Pattern 
10:15am - 10:45am Break 

10:45am - 11:30am

  • Arkadi Nemirovski 
  • Tight Semidefinite Relaxations and Statistical Estimation 

11:30am - 12:00pm

  • Sebastian Bubeck 
  • Metrical Task Systems on Trees 

12:00pm - 12:30pm

  • Raghu Pasupathy 
  • The Complexity of Adaptive Sampling Gradient and Adaptive Sampling 
Tuesday, May 15, 2018 
8:30-9:00 Continental Breakfast 

9:00am - 9:45am

  • Robert Nowak 
  • Outranked: Exploiting Nonlinear Algebraic Structure in Matrix Recovery Problems 

9:45am - 10:30am

  • Rachel Ward 
  • Learning the Learning Rate in Stochastic Gradient Descent 
10:30am - 11:00am Break 

11:00am - 11:45am

  • Alfred Hero 
  • Rate-Optimal Meta-Learning 

11:45am - 12:15pm

  • Yuejie Chi 
  • Geometry and Regularization in Nonconvex Statistical Estimation 
12:00pm - 2:00pm Lunch and POSTER SESSION 

2:15pm – 3:00pm

  • David Tse 
  • Understanding Generative Adversarial Networks 

3:00pm - 3:30pm

  • Sewoong Oh 
  • The Power of Two Samples for Generative Adversarial Networks 

3:30pm – 4:00pm

  • Wotao Yin 
  • R-Local Minimum Analysis and Run-And-Inspect Method for Certain Nonconvex Optimization with Complexity Guarantees 
4:00pm-4:30pm Break 

4:30pm - 5:00pm

  • Guanghui Lan 
  • Accelerated Stochastic Methods for Non-Convex Finite-Sum Optimization 

5:00pm - 5:30pm

  • John Wright 
  • Nonconvex Sparse Deconvolution: Geometry and Efficient Methods 

5:30pm - 6:15pm

  • Donald Goldfarb 
  • ADMM for Multiaffine Constrained Optimization: Theory and Applications