{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/evolutionary-learning-of-interpretable","title":"Evolutionary learning of interpretable decision trees","arxiv_id":"2012.07723","date":"2020-12-14","proceeding":null,"authors":["Leonardo Lucio Custode","Giovanni Iacca"],"abstract":"Reinforcement learning techniques achieved human-level performance in several tasks in the last decade. However, in recent years, the need for interpretability emerged: we want to be able to understand how a system works and the reasons behind its decisions. Not only we need interpretability to assess the safety of the produced systems, we also need it to extract knowledge about unknown problems. While some techniques that optimize decision trees for reinforcement learning do exist, they usually employ greedy algorithms or they do not exploit the rewards given by the environment. This means that these techniques may easily get stuck in local optima. In this work, we propose a novel approach to interpretable reinforcement learning that uses decision trees. We present a two-level optimization scheme that combines the advantages of evolutionary algorithms with the advantages of Q-learning. This way we decompose the problem into two sub-problems: the problem of finding a meaningful and useful decomposition of the state space, and the problem of associating an action to each state. We test the proposed method on three well-known reinforcement learning benchmarks, on which it results competitive with respect to the state-of-the-art in both performance and interpretability. Finally, we perform an ablation study that confirms that using the two-level optimization scheme gives a boost in performance in non-trivial environments with respect to a one-layer optimization technique.","url_abs":"https://arxiv.org/abs/2012.07723v3","url_pdf":"https://arxiv.org/pdf/2012.07723v3.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":"evolutionary-learning-of-interpretable","repo_url":"https://gitlab.com/leocus/ge_q_dts","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"evolutionary-algorithms","task_name":"Evolutionary Algorithms"},{"task_slug":"openai-gym","task_name":"OpenAI Gym"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[{"method_slug":"grammatical-evolution-q-learning","method_name":"Grammatical evolution + Q-learning"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/openai-gym-on-cartpole-v1","task":"OpenAI Gym","dataset":"CartPole-v1","model":"Orthogonal decision tree","rank_in_archive_order":1,"of":2,"metrics":{"Average Return":"500"},"uses_additional_data":false},{"leaderboard":"/sota/openai-gym-on-cartpole-v1","task":"OpenAI Gym","dataset":"CartPole-v1","model":"Oblique decision tree","rank_in_archive_order":2,"of":2,"metrics":{"Average Return":"500"},"uses_additional_data":false},{"leaderboard":"/sota/openai-gym-on-lunarlander-v2","task":"OpenAI Gym","dataset":"LunarLander-v2","model":"Oblique decision tree","rank_in_archive_order":1,"of":2,"metrics":{"Average Return":"272.14"},"uses_additional_data":false},{"leaderboard":"/sota/openai-gym-on-mountain-car","task":"OpenAI Gym","dataset":"Mountain Car","model":"Orthogonal decision tree","rank_in_archive_order":1,"of":2,"metrics":{"Average Return":"-101.72"},"uses_additional_data":false},{"leaderboard":"/sota/openai-gym-on-mountain-car","task":"OpenAI Gym","dataset":"Mountain Car","model":"Oblique decision tree","rank_in_archive_order":2,"of":2,"metrics":{"Average Return":"-106.02"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2012.07723","atlas_url":"https://app.syntology.ai/?focus=2012.07723","mcp":null,"developers":"https://syntology.ai/developers","held":true,"reason":"arxiv_id_in_atlas","checked_against":{"atlas_ids":549449,"atlas_ids_sha256":"afd72399dafd9906e9062927f70655952385a3d75488558e1f0484e950ab09b7","read_at":"2026-10-05T01:46:21Z"},"agent_calls":[{"tool":"get_citation_path","arguments":{"paper_1":"2012.07723"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2012.07723\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2012.07723"},"call":"get_concepts_for_paper(arxiv_id=\"2012.07723\")"}],"by_repo":[{"repo":"gitlab.com/leocus/ge_q_dts","kind":"official","samples":0,"ran":0,"instrument":0,"unverified":0}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code"},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"5c04aa0e8756770b93497c2e1d66740edfe5826e630c8ba6906a4517f4f79280","record_changed_at":"2026-09-28","record_changed_at_basis":"first_hashed"}