no code implementations • 4 Jun 2023 • Sam Powers, Abhinav Gupta, Chris Paxton
Robots in home environments need to be able to learn new skills continuously as data becomes available, becoming ever more capable over time while using as little real-world data as possible.
no code implementations • 21 Apr 2023 • Priyam Parashar, Vidhi Jain, Xiaohan Zhang, Jay Vakil, Sam Powers, Yonatan Bisk, Chris Paxton
We see a 4x improvement over baseline in mobile manipulation setting.
1 code implementation • 31 Dec 2022 • Sam Powers, Eliot Xing, Abhinav Gupta
The ability for an agent to continuously learn new skills without catastrophically forgetting existing knowledge is of critical importance for the development of generally intelligent agents.
2 code implementations • 19 Oct 2021 • Sam Powers, Eliot Xing, Eric Kolve, Roozbeh Mottaghi, Abhinav Gupta
In this work, we present CORA, a platform for Continual Reinforcement Learning Agents that provides benchmarks, baselines, and metrics in a single code package.