Online Decision Transformer
About
Recent work has shown that offline reinforcement learning (RL) can be formulated as a sequence modeling problem (Chen et al., 2021; Janner et al., 2021) and solved via approaches similar to large-scale language modeling. However, any practical instantiation of RL also involves an online component, where policies pretrained on passive offline datasets are finetuned via taskspecific interactions with the environment. We propose Online Decision Transformers (ODT), an RL algorithm based on sequence modeling that blends offline pretraining with online finetuning in a unified framework. Our framework uses sequence-level entropy regularizers in conjunction with autoregressive modeling objectives for sample-efficient exploration and finetuning. Empirically, we show that ODT is competitive with the state-of-the-art in absolute performance on the D4RL benchmark but shows much more significant gains during the finetuning procedure.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Locomotion | D4RL walker2d-medium-expert | Normalized Score3.04e+3 | 90 | |
| Locomotion | D4RL Walker2d medium | -- | 70 | |
| Locomotion | D4RL HalfCheetah Medium-Replay | -- | 68 | |
| Locomotion | D4RL hopper-medium-expert | -- | 28 | |
| Locomotion | D4RL hopper medium-replay | Test Return698.8 | 22 | |
| Locomotion | D4RL halfcheetah-medium-expert | Test Return3.57e+3 | 22 | |
| Locomotion | D4RL Cheetah Medium | Mean Return4.24e+3 | 17 | |
| Online Fine-tuning | D4RL MuJoCo and Maze2D online fine-tuning v2 v0 | Normalized Return97.5 | 14 | |
| Locomotion | D4RL walker2d medium-replay | Mean Return1.38e+3 | 12 |