CSAIL-MSR Trustworthy and Robust AI (TRAC) Workshop


Location:
32-D463 (Star) — Nov 14
32-G449 (Kiva/Patil) — Nov 15

Schedule

November 14, 2019 (32-D463 (Star))

8:30 - 9:00           Breakfast/Coffee
9:00 - 9:15 Welcome and Overview
Aleksander Madry (MIT), Evelyne Viegas (MSR)
9:15 - 10:00 Invited talk: Optimality and Approximation with Policy Gradient Methods in Markov Decision Processes
Alekh Agarwal (MSR)
10:00 - 10:30 Secure Distributed Machine Learning
Vinod Vaikuntanathan (MIT)
10:30 - 11:00 Mixed Autonomy Traffic: A Reinforcement Learning Perspective
Cathy Wu (MIT)
11:00 - 11:20 Coffee Break
11:20 - 11:50 Towards Safe Online Reinforcement Learning in Computer Systems
Mohammad Alizadeh (MIT)
11:50 - 12:20 Distributionally Robust Optimization and Generalization
Stefanie Jegelka (MIT)
12:20 - 13:20 Lunch
13:20 - 14:05 Invited Talk: A Program Synthesis Perspective on Explainable Machine Learning
Armando Solar-Lezama (MIT)
14:05 - 15:05 Lightning talks - Session I
Brandon Carter, Kai Xiao, Sitan Chen, Guang-He Lee, Yichen Yang
15:05 - 15:35 Approximate Cross Validation for Large Data and High Dimensions
Tamara Broderick (MIT)
15:35 - 16:00 Coffee Break
16:00 - 17:00 Panel: AI Robustness and Reliability
Tamara Broderick (MIT), Sertac Karaman (MIT), Besmira Nushi (MSR), Martin Rinard (MIT)
17:00 - 18:30 Reception

November 15, 2019 (32-G449 (Kiva/Patil))

8:30 - 9:00      Breakfast/Coffee                                                                                                                                  
9:00 - 9:45 Invited talk: Transparency and Intelligibility Throughout the Machine Learning Lifecycle
Jennifer Wortman Vaughan (MSR)
9:45 - 10:15 Efficient Federated Learning
Ankur Moitra (MIT)
10:15 - 10:45 Provable Pruning for Efficient and Interpretable Networks
Cenk Baykal (MIT)
10:45 - 11:05 Coffee Break
11:05 - 11:35 The Double-Edged Sword of ML
Aleksander Madry (MIT)
11:35 - 12:05 Robust Machine Learning for Reliable Patient Interventions
Maggie Makar (MIT)
12:05 - 13:05 Lunch
13:05 - 13:50 Invited Talk: Building AI We Can Trust by Reverse-Engineering Human Intelligence
Joshua Tenenbaum (MIT)
13:50 - 14:50 Lightning talks - Session II
Angie Boggust, Jose Javier Gonzalez, Alireza Fallah, Raj Agrawal, Lorenzo Masoero
14:50 - 15:20 Correctness Verification of Neural Networks
Martin Rinard (MIT)
15:20 - 15:45 Coffee Break
15:45 - 16:45 Panel: Human-AI Interaction
Sean Andrist (MSR), Arvind Satyanarayan (MIT), Antonio Torralba (MIT)
16:45 - 17:00 Closing Remarks
Eric Horvitz (MSR), Daniela Rus (MIT)