Overview

Tree search represents possible decisions as branches and allocates effort among them to find a path to a goal or a strong strategy. Methods include A* and best-first search with heuristics, Monte Carlo tree search with value estimates, bidirectional search, and learned policies that guide which branches to expand or prune. These approaches are used in motion and classical planning, games, optimization, and language-model agents that reason, use tools, or generate code. A central challenge is balancing broad exploration against costly evaluation and rollout budgets, while ensuring that pruning does not discard useful solutions or undermine guarantees.

Latest papers

CardsList

64 papers5 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.

  1. PuzzleJAX: A Benchmark for Reasoning and Learning

    Aug 22, 2025Sam Earle, Graham Todd, Yuchen Li +5RL BenchmarksGame-Playing Agents

  2. EcoSearch: A Constant-Delay Best-First Search Algorithm for Program Synthesis

    Dec 23, 2024Théo Matricon, Nathanaël Fijalkow, Guillaume LagardeProgram SynthesisTree Search

  3. Greedy Heuristics for Sampling-Based Motion Planning in High-Dimensional State Spaces

    May 6, 2024Phone Thiha Kyaw, Anh Vu Le, Rajesh Elara Mohan +1Sampling-Based Motion PlanningTree Search

  4. Solving Witness Triangle Puzzles Faster with an Automatically Learned Predicate

    Aug 4, 2023Justin Stevens, Vadim Bulitko, David ThueAutomated Heuristic DesignInference-Time Search