AI Paper Atlas

Research map / Learning theory and algorithms / Combinatorial and search optimization

Single-Agent Policy Tree Search With Guarantees

Laurent Orseau, Levi Lelis, Tor Lattimore, Theophane Weber

NeurIPS 2018 · Learning theory and algorithms › Combinatorial and search optimization · approach: Monte Carlo tree search · 18 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
Use LevinTS guided by an A3C policy instead of LAMASokoban solver · Sokoban, 1,000 levelsAvg. length: 51.6 -> 39.8Improved
Use LevinTS guided by an A3C policy instead of LAMASokoban solver · Sokoban, 1,000 levelsTotal expansions: 3,151,325 -> 6,602,666Hurt
Add 1% noise to the policy used by LevinTS (ablation)LevinTS · Sokoban, 1,000 levelsTotal expansions: 6,602,666 -> 5,026,200Improved
Add 1% noise to the policy used by LevinTS (ablation)LevinTS · Sokoban, 1,000 levelsSolved: 1,000 -> 1,000No clear effect
Use LubyTS(256, 32) instead of LubyTS(256, 1) (ablation)LubyTS · Sokoban, 1,000 levelsSolved: 753 ± 5 -> 870 ± 2Improved
Use LubyTS(256, 32) instead of LubyTS(256, 1) (ablation)LubyTS · Sokoban, 1,000 levelsTotal expansions: 63,8481 ± 2,434 -> 6,246,293 ± 73,382Mixed
Add 1% noise to the policy used by LubyTS(512, 32) (ablation)LubyTS(512, 32) · Sokoban, 1,000 levelsSolved: 884 ± 4 -> 896 ± 2Improved
Add 1% noise to the policy used by MultiTS(200, 200) (ablation)MultiTS(200, 200) · Sokoban, 1,000 levelsSolved: 881 ± 1 -> 895 ± 3Improved
Use MultiTS(200, 200) instead of MultiTS(1, 200) (ablation)MultiTS · Sokoban, 1,000 levelsSolved: 669 ± 5 -> 881 ± 1Improved
Use policy-guided search instead of sampling a single trajectoryPolicy tree search · Sokoban, 1,000 levelsSolved: The text reports more than 20% improvementImproved

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