June 15, 2023
Model-based reinforcement learning (RL) has shown great promise due to its sample efficiency, but still struggles with long-horizon sparse-reward tasks, especially in offline settings where the agent learns from a fixed dataset. We hypothesize that model-based RL agents struggle in these environments due to a lack of long-term planning capabilities, and that planning in a temporally abstract model of the environment can alleviate this issue. In this paper, we make two key contributions: 1) we introduce an offline model-based RL algorithm, IQL-TD-MPC, that extends the state-of-the-art Temporal Difference Learning for Model Predictive Control (TD-MPC) with Implicit Q-Learning (IQL); 2) we propose to use IQL-TD-MPC as a Manager in a hierarchical setting with any off-the-shelf offline RL algorithm as a Worker. More specifically, we pre-train a temporally abstract IQL-TD-MPC Manager to predict "intent embeddings", which roughly correspond to subgoals, via planning. We empirically show that augmenting state representations with intent embeddings generated by an IQL-TD-MPC manager significantly improves off-the-shelf offline RL agents' performance on some of the most challenging D4RL benchmark tasks. For instance, the offline RL algorithms AWAC, TD3-BC, DT, and CQL all get zero or near-zero normalized evaluation scores on the medium and large antmaze tasks, while our modification gives an average score over 40.
Written by
Bobak Hashemi
Lucas Lehnert
Rohan Chitnis
Urun Dogan
Olivier Delalleau
Publisher
Arxiv
Research Topics
July 01, 2024
Andrei Lupu, Chris Lu, Robert Lange, Jakob Foerster
July 01, 2024
May 06, 2024
Haoyue Tang, Tian Xie
May 06, 2024
April 30, 2024
Mikayel Samvelyan, Minqi Jiang, Davide Paglieri, Jack Parker-Holder, Tim Rocktäschel
April 30, 2024
April 02, 2024
Patrick Lancaster, Nicklas Hansen, Aravind Rajeswaran, Vikash Kumar
April 02, 2024
Foundational models
Latest news
Foundational models