{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:OLU234XGMVAXCPLOFTPHKLSXY5","short_pith_number":"pith:OLU234XG","schema_version":"1.0","canonical_sha256":"72e9adf2e66541713d6e2cde752e57c75ea4ee02c6cdae707c1ed5bff9168459","source":{"kind":"arxiv","id":"2204.03771","version":1},"attestation_state":"computed","paper":{"title":"Q-learning with online random forests","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Joosung Min, Lloyd T. Elliott","submitted_at":"2022-04-07T23:00:39Z","abstract_excerpt":"$Q$-learning is the most fundamental model-free reinforcement learning algorithm. Deployment of $Q$-learning requires approximation of the state-action value function (also known as the $Q$-function). In this work, we provide online random forests as $Q$-function approximators and propose a novel method wherein the random forest is grown as learning proceeds (through expanding forests). We demonstrate improved performance of our methods over state-of-the-art Deep $Q$-Networks in two OpenAI gyms (`blackjack' and `inverted pendulum') but not in the `lunar lander' gym. We suspect that the resilie"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2204.03771","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"stat.ML","submitted_at":"2022-04-07T23:00:39Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c29537f97e04f3698fb95c79988604d2c9f2c157bdab824223181c79adcc6e54","abstract_canon_sha256":"2722ed711f57faa4c5e4e28b8a59908c25763fce86ff734cf1e1df78b465435c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:12:41.690242Z","signature_b64":"/9SLZXCtqEeEqDnS8zI54RsKjRNEwlgYGmytdSBtlGEvA/nV+gemErkDR3giiby0Gz58thEZ6eb6qqzs2c0iAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"72e9adf2e66541713d6e2cde752e57c75ea4ee02c6cdae707c1ed5bff9168459","last_reissued_at":"2026-07-05T04:12:41.689879Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:12:41.689879Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Q-learning with online random forests","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Joosung Min, Lloyd T. Elliott","submitted_at":"2022-04-07T23:00:39Z","abstract_excerpt":"$Q$-learning is the most fundamental model-free reinforcement learning algorithm. Deployment of $Q$-learning requires approximation of the state-action value function (also known as the $Q$-function). In this work, we provide online random forests as $Q$-function approximators and propose a novel method wherein the random forest is grown as learning proceeds (through expanding forests). We demonstrate improved performance of our methods over state-of-the-art Deep $Q$-Networks in two OpenAI gyms (`blackjack' and `inverted pendulum') but not in the `lunar lander' gym. We suspect that the resilie"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.03771","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2204.03771/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2204.03771","created_at":"2026-07-05T04:12:41.689942+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.03771v1","created_at":"2026-07-05T04:12:41.689942+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.03771","created_at":"2026-07-05T04:12:41.689942+00:00"},{"alias_kind":"pith_short_12","alias_value":"OLU234XGMVAX","created_at":"2026-07-05T04:12:41.689942+00:00"},{"alias_kind":"pith_short_16","alias_value":"OLU234XGMVAXCPLO","created_at":"2026-07-05T04:12:41.689942+00:00"},{"alias_kind":"pith_short_8","alias_value":"OLU234XG","created_at":"2026-07-05T04:12:41.689942+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.11457","citing_title":"Upside-Down Reinforcement Learning for More Interpretable Optimal Control","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5","json":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5.json","graph_json":"https://pith.science/api/pith-number/OLU234XGMVAXCPLOFTPHKLSXY5/graph.json","events_json":"https://pith.science/api/pith-number/OLU234XGMVAXCPLOFTPHKLSXY5/events.json","paper":"https://pith.science/paper/OLU234XG"},"agent_actions":{"view_html":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5","download_json":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5.json","view_paper":"https://pith.science/paper/OLU234XG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.03771&json=true","fetch_graph":"https://pith.science/api/pith-number/OLU234XGMVAXCPLOFTPHKLSXY5/graph.json","fetch_events":"https://pith.science/api/pith-number/OLU234XGMVAXCPLOFTPHKLSXY5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5/action/storage_attestation","attest_author":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5/action/author_attestation","sign_citation":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5/action/citation_signature","submit_replication":"https://pith.science/pith/OLU234XGMVAXCPLOFTPHKLSXY5/action/replication_record"}},"created_at":"2026-07-05T04:12:41.689942+00:00","updated_at":"2026-07-05T04:12:41.689942+00:00"}