AI Paper Atlas

Research map / Reinforcement learning / Visual and representation RL

Ranking Policy Decisions

Hadrien Pouget, Hana Chockler, Youcheng Sun, Daniel Kroening

NeurIPS 2021 · Reinforcement learning › Visual and representation RL · approach: Probabilistic and autonomous RL · 1 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
Rank decisions with the SBFL portfolio instead of random rankingPruned policies using repeat-previous-action default · MiniGrid% of original states restored to recover 90% performance: 99 \pm 00 -> 49 \pm 00Improved
Rank decisions with the SBFL portfolio instead of random rankingPruned policies using repeat-previous-action default · CartPole% of original states restored to recover 90% performance: 65 \pm 02 -> 31 \pm 04Improved
Rank decisions with the SBFL portfolio instead of random rankingPruned policies using repeat-previous-action default · Breakout% of original states restored to recover 90% performance: 100 \pm 00 -> 10 \pm 00Improved
Rank decisions with the SBFL portfolio instead of random rankingPruned policies using repeat-previous-action default · Pong% of original states restored to recover 90% performance: 79 \pm 03 -> 21 \pm 03Improved
Use state-visit frequency instead of SBFL to rank decisionsPolicy pruning · Benchmarks reported in the paperpruned-policy performance per original-policy step usageHurt
Replace repeat-previous-action default with random actions (ablation)SBFL-ranked pruned policies · Benchmarks with both default-action experimentspruning efficiency at recovering original performance: In most games, the difference was small (\leq 10\%); a few games show a marginal improvement.Mixed
Use a specialized ball-and-paddle coordinate abstraction instead of the generic Breakout abstraction (ablation)Breakout SBFL ranking · Breakout% of steps using original policy to recover 90% performance: 13\% fewer stepsImproved
Use a specialized ball-and-paddle coordinate abstraction instead of the generic Breakout abstraction (ablation)Breakout SBFL ranking · Breakout% of original states restored to recover 90% performanceHurt

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