{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/single-agent-policy-tree-search-with","title":"Single-Agent Policy Tree Search With Guarantees","arxiv_id":"1811.10928","date":"2018-11-27","proceeding":"NeurIPS 2018 12","authors":["Laurent Orseau","Levi H. S. Lelis","Tor Lattimore","Théophane Weber"],"abstract":"We introduce two novel tree search algorithms that use a policy to guide\nsearch. The first algorithm is a best-first enumeration that uses a cost\nfunction that allows us to prove an upper bound on the number of nodes to be\nexpanded before reaching a goal state. We show that this best-first algorithm\nis particularly well suited for `needle-in-a-haystack' problems. The second\nalgorithm is based on sampling and we prove an upper bound on the expected\nnumber of nodes it expands before reaching a set of goal states. We show that\nthis algorithm is better suited for problems where many paths lead to a goal.\nWe validate these tree search algorithms on 1,000 computer-generated levels of\nSokoban, where the policy used to guide the search comes from a neural network\ntrained using A3C. Our results show that the policy tree search algorithms we\nintroduce are competitive with a state-of-the-art domain-independent planner\nthat uses heuristic search.","url_abs":"http://arxiv.org/abs/1811.10928v2","url_pdf":"http://arxiv.org/pdf/1811.10928v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"single-agent-policy-tree-search-with","repo_url":"https://github.com/deepmind/boxoban-levels","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"heuristic-search","task_name":"Heuristic Search"},{"task_slug":"sokoban","task_name":"Sokoban"}],"methods":[{"method_slug":"a3c","method_name":"A3C"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"entropy-regularization","method_name":"Entropy Regularization"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.10928","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}