Skip to main content
Search

# DelayBench

1 related articles

Subscribe to updates
  1. arXiv

    Training agents under delayed observations: Hindsight Critics changes only the critics and tops delay robustness in 4 of 8 environments

    Observation delays leave an agent acting on a past state, and standard reinforcement learning degrades as a result. Hindsight Critics estimates the current state from completed trajectories during training and gives that estimate only to the critics, while the actor still sees delayed observations at deployment. On the 8 continuous-control environments of DelayBench, it posts the highest delay-robustness score in 4, partially offsetting moderate delays, though delays of 16 timesteps remain hard.

You’re all caught up