Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Bootstrapped Transformer for Offline Reinforcement Learning

About

Offline reinforcement learning (RL) aims at learning policies from previously collected static trajectory data without interacting with the real environment. Recent works provide a novel perspective by viewing offline RL as a generic sequence generation problem, adopting sequence models such as Transformer architecture to model distributions over trajectories, and repurposing beam search as a planning algorithm. However, the training datasets utilized in general offline RL tasks are quite limited and often suffer from insufficient distribution coverage, which could be harmful to training sequence generation models yet has not drawn enough attention in the previous works. In this paper, we propose a novel algorithm named Bootstrapped Transformer, which incorporates the idea of bootstrapping and leverages the learned model to self-generate more offline data to further boost the sequence model training. We conduct extensive experiments on two offline RL benchmarks and demonstrate that our model can largely remedy the existing offline RL training limitations and beat other strong baseline methods. We also analyze the generated pseudo data and the revealed characteristics may shed some light on offline RL training. The codes are available at https://seqml.github.io/bootorl.

Kerong Wang, Hanye Zhao, Xufang Luo, Kan Ren, Weinan Zhang, Dongsheng Li• 2022

Related benchmarks

TaskDatasetResultRank
Offline Reinforcement LearningD4RL walker2d-random
Normalized Score5.2
77
Offline Reinforcement LearningD4RL halfcheetah-random
Normalized Score7.5
70
Offline Reinforcement LearningD4RL hopper-random
Normalized Score6.8
62
Hand ManipulationAdroit door-human
Normalized Avg Score0.6
33
Hand ManipulationAdroit door-cloned
Normalized Score0.00e+0
23
Offline Reinforcement LearningD4RL Walker2d expert
Mean Normalized Score108.7
22
Hand ManipulationAdroit door-expert
Normalized Avg Score106.5
21
Offline Reinforcement LearningD4RL Gym v2 (test)
Score (HalfCheetah, Medium-Expert)94
20
Hand ManipulationAdroit pen-human
Normalized Average Score54.3
19
Offline Reinforcement LearningD4RL Hopper (expert)
Mean Normalized Score110.5
16
Showing 10 of 19 rows

Other info

Code

Follow for update