AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Understanding and Preventing Capacity Loss in Reinforcement Learning

Clare Lyle, Mark Rowland, Will Dabney

ICLR 2022 · Reinforcement learning › Visual and representation RL · approach: In-context and adaptive RL · 10 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
Train on a sequence of changing target functions instead of a single targetNeural networks on MNIST and value-based RL agents · Non-stationary target-fitting taskstarget-fitting errorHurt
Add Initial Feature Regularization (InFeR)Rainbow · Atari 57-game benchmarkhuman-normalized scoreImproved
Add InFeRRainbow · Montezuma's RevengeperformanceImproved
Add InFeRDouble DQN · Montezuma's RevengeperformanceImproved
Add InFeRDouble DQN · Atari benchmarkaverage human-normalized scoreImproved
Add InFeRRainbow · Asteroids and JamesbondperformanceHurt
Add InFeR to the regression objectiveFully connected network · Non-stationary MNIST random reward predictiontarget-fitting capacityImproved
Add InFeRRainbow · Atari gamesfeature rankImproved
Add random features to the learned representation instead of using InFeR (ablation)Rainbow · Atari gamesperformanceNo clear effect
Double the width of the penultimate layer while using InFeR (ablation)Rainbow · Games where InFeR hurt in the original networkperformance gap versus RainbowImproved
Use a random cumulant prediction auxiliary taskDQN · Montezuma's RevengeperformanceImproved
Use a random cumulant prediction auxiliary taskDQN · Complex dense-reward gameslearning progressHurt

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