RT augments offline RL datasets with backward-generated, reliability-truncated, high-reward-conditioned trajectories.
De- cision transformer: Reinforcement learning via sequence modeling
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Model-Based Offline Reinforcement Learning with Reliability-Guaranteed Sequence Modeling
RT augments offline RL datasets with backward-generated, reliability-truncated, high-reward-conditioned trajectories.