Intersections between Control, Learning and Optimization - IPAM

Intersections between Control, Learning and Optimization

February 24 - 28, 2020

Schedule

All times in this Schedule are Pacific Time (PT)

Monday, February 24, 2020

Morning Session

08:00-08:55
Check-In/Breakfast (Hosted by IPAM)
08:55-09:00
Welcome & Opening Remarks: Dean Miguel García-Garibay (Dean of Physical Sciences, UCLA) and Dima Shlyakhtenko (Director, IPAM)
09:00-09:50
Moritz Diehl (University of Freiburg)
Convexity Exploiting Newton-Type Optimization for Learning and Control
10:00-10:15
Break
10:15-11:05
Dimitri Bertsekas (Massachusetts Institute of Technology and Arizona State University)
Distributed and Multiagent Reinforcement Learning
11:15-11:30
Break
11:30-12:20
Csaba Szepesvari (University of Alberta)
Model misspecification in reinforcement learning

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Necmiye Ozay (University of Michigan)
A fresh look at some classical system identification methods
15:30-16:00
Break
16:00-16:50
Mengdi Wang (Princeton University)
On the statistical complexity of reinforcement learning
17:00-18:30
Poster Session & Reception (Hosted by IPAM)

Tuesday, February 25, 2020

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:50
Russell Tedrake (Massachusetts Institute of Technology)
From pixels to torques: output feedback for robotics
10:00-10:15
Break
10:15-11:05
Martin Riedmiller (DeepMind Technologies)
Learning Control from Minimal Prior Knowledge
11:15-11:30
Break
11:30-12:20
Ben Recht (University of California, Berkeley (UC Berkeley))
Trying to Make Sense of Control from Pixels

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Emma Brunskill (Stanford University)
Curbing Our Enthusiasm: Constraining Decision Policies Learned from the Past to Ensure Good Futures
15:30-16:00
Break
16:00-16:50
Anca Dragan (University of California, Berkeley (UC Berkeley))
Learning Intended Cost Functions: Extracting all the right information from all the right places

Wednesday, February 26, 2020

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:50
Andreas Krause (ETH Zurich)
Safe and Efficient Exploration in Reinforcement Learning
10:00-10:15
Break
10:15-11:05
Daniel Kuhn (École Polytechnique Fédérale de Lausanne (EPFL))
Wasserstein Distributionally Robust Optimization: Theory and Applications in Machine Learning
11:15-11:30
Break
11:30-12:20
Zico Kolter (Carnegie Mellon University)
Integrating optimization, constraints, and control within deep learning models

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
Melanie Zeilinger (ETH Zurich)
Learning-based Model Predictive Control - Towards Safe Learning in Control
15:30-16:00
Break
16:00-16:50
Francesco Borrelli (University of California, Berkeley (UC Berkeley))
Sample-Based Learning Model Predictive Control

Thursday, February 27, 2020

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:50
Maryam Fazel (University of Washington)
Finite-sample System Identification: Optimal Rates and the Role of Regularization
10:00-10:15
Break
10:15-11:05
Richard Murray (California Institute of Technology)
Can We Really Use Machine Learning in Safety Critical Systems?
11:15-11:30
Break
11:30-12:20
Dorsa Sadigh (Stanford University)
Beyond Theory of Mind: Learning and Influencing Conventions in Multi-Agent Interactions

Afternoon Session

12:30-14:30
Lunch (on your own)
14:30-15:20
James Rawlings (University of California, Santa Barbara (UCSB))
Industrial, large-scale model predictive control with deep neural networks
15:30-16:00
Break
16:00-16:50
Stephen Wright (University of Wisconsin-Madison)
Nonconvex optimization in matrix optimization and distributionally robust optimization

Friday, February 28, 2020

Morning Session

08:00-09:00
Check-In/Breakfast (Hosted by IPAM)
09:00-09:10
Rohit Kannan (University of Wisconsin-Madison)
Predict, then smart optimize with stochastic programming
09:15-09:25
Sungho Shin (University of Wisconsin-Madison)
A Unifying Framework for Subspace Identification and Dynamic Mode Decomposition
09:30-09:40
Victor Magron (Laboratoire d'analyse et d'architecture des systèmes (LAAS-CNRS))
Polynomial Optimization for Bounding Lipschitz Constants of Deep Networks
09:45-09:55
Jia-Jie Zhu (Max Planck Institute for Intelligent Systems)
Distributionally Robust Optimization and Control using RKHS Embedding
10:00-10:15
Break
10:15-11:05
Lieven Vandenberghe (University of California, Los Angeles (UCLA))
Bregman proximal methods for semidefinite optimization.
11:15-11:30
Break
11:30-12:20
Angela Schoellig (University of Toronto)
Machine Learning for Robotics: Achieving Safety, Performance and Reliability by Combining Models and Data in a Closed-Loop System Architecture

Afternoon Session

12:30-12:30
Conclusion
Printable Version