Timezone: »

Bootstrapped Transformer for Offline Reinforcement Learning
Kerong Wang · Hanye Zhao · Xufang Luo · Kan Ren · Weinan Zhang · Dongsheng Li

Wed Nov 30 02:00 PM -- 04:00 PM (PST) @ Hall J #202

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 suffering from insufficient distribution coverage, which could me 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 training of sequence model. We conduct extensive experiments on two offline RL benchmarks and demonstrate that our model can largely remedy the limitations of the existing offline RL training 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.

Author Information

Kerong Wang (Shanghai Jiaotong University)
Hanye Zhao (Shanghai Jiao Tong University)
Xufang Luo (Microsoft Research)
Kan Ren (Microsoft)
Weinan Zhang (Shanghai Jiao Tong University)
Dongsheng Li (IBM Research - China)

More from the Same Authors