Robotic AI & Learning Lab



Welcome to the RAIL lab website! Our research focus is to enable machines to exhibit flexible and adaptable behavior, acquired autonomously through learning. To that end, we work on learning algorithms, robotics, and computer vision.

News

December 13, 2019 Find out how CycleGAN, together with visual model-based RL, can allow robots to imitate videos of humans by directly translating videos, pixel by pixel, in a new BAIR blog post by Laura Smith and Marvin Zhang!
December 12, 2019 Michael Janner summarizes the state of the art in model-based RL and describes model-based policy optimization (MBPO) in his new BAIR blog post!
December 5, 2019 Data-driven deep reinforcement learning -- offline RL that learns from data. See the new blog post from Aviral Kumar!
November 26, 2019 A new BAIR blog post by Sudeep Dasari on RoboNet, a large dataset of multi-robot interaction data, is now online!
September 30, 2019 A new BAIR blog post by Anusha Nagabandi on our work on model-based RL for dexterous manipulation is now online!
June 21, 2019 An excellent short talk by Vitchyr Pong covering the Skew-Fit paper is now available online!
June 19, 2019 An excellent short talk by Michael Janner on the paper "When to Trust Your Model," covering the model-based policy optimization (MBPO) algorithm, is now available online!
May 15, 2019 Dr. Chelsea Finn, former RAIL Ph.D. student and current post-doc, wins the 2019 ACM Dissertation Award! Congratulations Chelsea!
April 11, 2019 Our research on robots that learn how to use tools has been featured in MIT Technology Review! Congratulations to all the authors, especially the lead author Annie Xie!
January 9, 2019 Our research on soft actor-critic algorithms for robotic walking has been featured on WIRED! Congratulations to all the authors, especially the lead author Tuomas Haarnoja!
December 18, 2018 Congratulations to Dr. Tuomas Haarnoja for completing his Ph.D.!
December 14, 2018 BAIR blog post on our work on sample-efficient deep RL: Soft Actor Critic - Deep Reinforcement Learning with Real-World Robots
November 30, 2018 BAIR blog post on our work on model-based RL from pixels for real-world robotic control: Visual Model-Based Reinforcement Learning as a Path towards Generalist Robots
October 9, 2018 BAIR blog post on our SIGGRAPH Asia 2018 paper: Learning Acrobatics by Watching YouTube
September 6, 2018 BAIR blog post on our vision-based RL work: Visual Reinforcement Learning with Imagined Goals
August 31, 2018 BAIR blog post on our dexterous manipulation work: Dexterous Manipulation with Reinforcement Learning: Efficient, General, and Low-Cost
August 17, 2018 Congratulations to Dr. Chelsea Finn for completing her Ph.D.!
June 28, 2018 BAIR blog post on our RSS 2018 paper: One-Shot Imitation from Watching Videos
April 20, 2018 BAIR blog post on our SIGGRAPH 2018 paper: Towards a Virtual Stuntman
April 19, 2017 BAIR blog post published about one of our RSS 2018 papers: Shared Autonomy via Deep Reinforcement Learning
January 29, 2017 Seven papers accepted at the 2018 International Conference on Learning Representations (ICLR), see the publications page. Preprints for all papers coming soon!
January 12, 2017 Ten papers accepted at the 2018 International Conference on Robotics and Automation (ICRA), see the publications page. Preprints for all papers coming soon!
December 5, 2017 We showcased some of our research in a live robot demo at NIPS 2017. See here for more information.

Research Support

National Science Foundation, 2016 - present
Google, 2016 - present
Honda, 2017 - present
Berkeley DeepDrive (BDD), 2016 - present
Open Philanthropy Project, 2017 - present
Office of Naval Research, 2016 - present
NVIDIA, 2016 - present
Berkeley AI Research (BAIR), 2016 - present