L4DC: Learning for Dynamics and Control
Kirsch Auditorium (32-123)
Thursday, May 30 & Friday, May 31, 2019
Thursday, May 30, 2019
Morning Session
08:15 – 08:45: Registration
08:45-09:00: Welcome and Introduction by organizers
09:00-09:30: A Practitioner’s Perspective, Manfred Morari (University of Pennsylvania)
09:30-10:00: Learning to Predict and Control, Francesco Borrelli (UC Berkeley)
10:00-10:30: Machine Learning for Robotics: Safety and Performance Guarantees for Learning-Based Control, Angela Schoellig (University of Toronto)
10:30-11:00: Coffee Break
11:00-11:30: Gradient based methods for linear system control and identification, Maryam Fazel (University of Washington)
11:30-12:00: Dynamics and Control of Differential Learning, Stefano Soatto (UCLA)
12:00 – 14:00: Lunch Break and Poster Session
Afternoon Session
14:00-14:30: An Optimization-Centric Theory of Mind for Human-Robot Interaction, Anca Dragan (UC Berkeley)
14:30-15:00: Integrating Autonomy into Urban Systems, Cathy Wu (MIT/Microsoft Research)
15:00-16:00: Planning and Discussion
16:00-16:30: Coffee Break
16:30-17:00: Efficient Reinforcement Learning When Data is Costly, Emma Brunskill (Stanford University)
17:00-17:30: Safe Exploration in Reinforcement Learning, Andreas Krause (ETH Zuerich)
17:30-18:30: Reception at R&D Commons
19:00-21:00: Banquet Dinner at Catalyst Restaurant (by invitation only)
Friday, May 31, 2019
Morning Session
09:00-09:30: Safe Learning in Robotics, Claire Tomlin (UC Berkeley)
09:30-10:00: Influencing Interactive Mixed-Autonomy Systems, Dorsa Sadigh (Stanford University)
10:00-10:30: From Optimization Algorithms to Continuous Dynamical Systems and Back, Rene Vidal (Johns Hopkins University)
10:30-11:00: Coffee Break
11:00-11:30: Acceleration-based methods for trajectory optimization through contacts, Emo Todorov (University of Washington)
11:30-12:00: Learning manipulation — and why I (still) like F=ma, Russ Tedrake (MIT)
12:00-14:00: Lunch Break and Poster Session
Afternoon Session
14:00-14:30: The Shades of Reinforcement Learning, John Tsitsiklis (MIT)
14:30-15:00: Robots That Learn By Doing, Sergey Levine (UC Berkeley)
15:00-15:30: A No Regret Algorithm for Robust Online Adaptive Control, Sham Kakade (University of Washington)
15:30-16:00: Coffee Break
16:00-16:30: Infusing Physics and Structure into Machine Learning, Anima Anandkumar (CalTech)
16:30-17:00: Cheating with neural networks: How to make the networks you hate obey the constraints you love, Zico Kolter (Carnegie Mellon University)
17:00: Adjourn
