{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:3HDAWHLINYZKNGLMGSCBBO3MLR","short_pith_number":"pith:3HDAWHLI","schema_version":"1.0","canonical_sha256":"d9c60b1d686e32a6996c348410bb6c5c6789eb6f0b20c606fd4e9503ad99baf4","source":{"kind":"arxiv","id":"2010.05545","version":1},"attestation_state":"computed","paper":{"title":"Local Search for Policy Iteration in Continuous Control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abbas Abdolmaleki, Arunkumar Byravan, Dan Belov, Daniel Mankowitz, Jackie Kay, Jonas Buchli, Jonas Degrave, Josh Merel, Jost Tobias Springenberg, Julian Schrittwieser, Martin Riedmiller, Nicolas Heess, Yuval Tassa","submitted_at":"2020-10-12T09:02:48Z","abstract_excerpt":"We present an algorithm for local, regularized, policy improvement in reinforcement learning (RL) that allows us to formulate model-based and model-free variants in a single framework. Our algorithm can be interpreted as a natural extension of work on KL-regularized RL and introduces a form of tree search for continuous action spaces. We demonstrate that additional computation spent on model-based policy improvement during learning can improve data efficiency, and confirm that model-based policy improvement during action selection can also be beneficial. Quantitatively, our algorithm improves "},"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":"2010.05545","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-12T09:02:48Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"a8515aa07884dd7ad05080ec0c45757abef83cecdda3a257d7d43aa70fa3f97b","abstract_canon_sha256":"419fe29c702073cc9556a61a0d81c310ea23c491722558f988258c57c2e2e368"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:42:07.761164Z","signature_b64":"kMkiUQzgq7eHUuv8fMrL8jhSrP8R3cPiBreGPLr4ufHAspktJY+E10RjDeWIVg/CorfsHHXNxvZCl1h77DcPAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d9c60b1d686e32a6996c348410bb6c5c6789eb6f0b20c606fd4e9503ad99baf4","last_reissued_at":"2026-07-05T01:42:07.760768Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:42:07.760768Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Local Search for Policy Iteration in Continuous Control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abbas Abdolmaleki, Arunkumar Byravan, Dan Belov, Daniel Mankowitz, Jackie Kay, Jonas Buchli, Jonas Degrave, Josh Merel, Jost Tobias Springenberg, Julian Schrittwieser, Martin Riedmiller, Nicolas Heess, Yuval Tassa","submitted_at":"2020-10-12T09:02:48Z","abstract_excerpt":"We present an algorithm for local, regularized, policy improvement in reinforcement learning (RL) that allows us to formulate model-based and model-free variants in a single framework. Our algorithm can be interpreted as a natural extension of work on KL-regularized RL and introduces a form of tree search for continuous action spaces. We demonstrate that additional computation spent on model-based policy improvement during learning can improve data efficiency, and confirm that model-based policy improvement during action selection can also be beneficial. Quantitatively, our algorithm improves "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.05545","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/2010.05545/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":"2010.05545","created_at":"2026-07-05T01:42:07.760826+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.05545v1","created_at":"2026-07-05T01:42:07.760826+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.05545","created_at":"2026-07-05T01:42:07.760826+00:00"},{"alias_kind":"pith_short_12","alias_value":"3HDAWHLINYZK","created_at":"2026-07-05T01:42:07.760826+00:00"},{"alias_kind":"pith_short_16","alias_value":"3HDAWHLINYZKNGLM","created_at":"2026-07-05T01:42:07.760826+00:00"},{"alias_kind":"pith_short_8","alias_value":"3HDAWHLI","created_at":"2026-07-05T01:42:07.760826+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR","json":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR.json","graph_json":"https://pith.science/api/pith-number/3HDAWHLINYZKNGLMGSCBBO3MLR/graph.json","events_json":"https://pith.science/api/pith-number/3HDAWHLINYZKNGLMGSCBBO3MLR/events.json","paper":"https://pith.science/paper/3HDAWHLI"},"agent_actions":{"view_html":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR","download_json":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR.json","view_paper":"https://pith.science/paper/3HDAWHLI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.05545&json=true","fetch_graph":"https://pith.science/api/pith-number/3HDAWHLINYZKNGLMGSCBBO3MLR/graph.json","fetch_events":"https://pith.science/api/pith-number/3HDAWHLINYZKNGLMGSCBBO3MLR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR/action/storage_attestation","attest_author":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR/action/author_attestation","sign_citation":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR/action/citation_signature","submit_replication":"https://pith.science/pith/3HDAWHLINYZKNGLMGSCBBO3MLR/action/replication_record"}},"created_at":"2026-07-05T01:42:07.760826+00:00","updated_at":"2026-07-05T01:42:07.760826+00:00"}