REVIEW 2 cited by
Consistent Dropout for Policy Gradient Reinforcement Learning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Dropout has long been a staple of supervised learning, but is rarely used in reinforcement learning. We analyze why naive application of dropout is problematic for policy-gradient learning algorithms and introduce consistent dropout, a simple technique to address this instability. We demonstrate consistent dropout enables stable training with A2C and PPO in both continuous and discrete action environments across a wide range of dropout probabilities. Finally, we show that consistent dropout enables the online training of complex architectures such as GPT without needing to disable the model's native dropout.
Forward citations
Cited by 2 Pith papers
-
Generalized Locomotion in Out-of-distribution Conditions with Robust Transformer
A transformer with body tokenization and consistent dropout generalizes to unseen leg damages and sensor noise while trained on limited dynamics and clean observations.
-
Spatially-Enhanced Recurrent Memory for Long-Range Mapless Navigation via End-to-End Reinforcement Learning
A modified recurrent unit with an input-multiplied gate improves spatial memory and long-range mapless navigation success rates by about 23.5% over standard RNNs in simulation and transfers zero-shot to a real robot.
Discussion (0). Sign in to comment.