AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Contrastive Retrospection: honing in on critical steps for rapid learning and generalization in RL

Chen Sun, Wannan Yang, Thomas Jiralerspong, Dane Malenfant, Benjamin Alsbury-Nealy, Yoshua Bengio, Blake Aaron Richards

NeurIPS 2023 · Reinforcement learning › Visual and representation RL · approach: Probabilistic and autonomous RL · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add ConSpec intrinsic reward to PPOPPO · 3D OrangeTreeepisode rewardImproved
add Synthetic Returns to PPOPPO · 3D OrangeTreeepisode rewardNo clear effect
add CURL to PPOPPO · 3D OrangeTreeepisode rewardNo clear effect
test with never-seen black flat objects (zero-shot)PPO+ConSpec · OrangeTree OoD colourepisode rewardImproved
test in a new room with different wall and floor colours after at most 20 PPO gradient stepsPPO+ConSpec · OrangeTree OoD roomepisode rewardImproved
increase number of keys in multi-key-to-door taskSynthRs, TVT, RUDDER, RND · multi-key-to-doorsuccess rateHurt
increase number of keys in multi-key-to-door taskPPO+ConSpec · multi-key-to-doortraining time excluding explorationNo clear effect
use 3 prototypes instead of 4 in the 4-key task (ablation)PPO+ConSpec · 4-key-to-doorsuccess rate: 4 -> 3 prototypesNo clear effect
remove prototype freezing (ablation)PPO+ConSpec · multi-key-to-doorsuccess rateNo clear effect
use policy-invariant intrinsic reward instead of simple max-score rewardPPO+ConSpec · multi-key-to-doorsuccess rateNo clear effect
change intrinsic reward scale lambda from 0.2 to 0.5 (ablation)PPO+ConSpec · OrangeTree, 4-key, Montezuma's Revengeepisode reward: 0.2 -> 0.5No clear effect
add ConSpec to PPO on delayed AtariPPO · delayed Pong, Bowling, AirRaidrewardImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy