{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:Y4NV6U3H72NMR5RV5ZOACHAUN6","short_pith_number":"pith:Y4NV6U3H","schema_version":"1.0","canonical_sha256":"c71b5f5367fe9ac8f635ee5c011c146f98cac13e4ba2ce6ef7dcbc1ab4b3760a","source":{"kind":"arxiv","id":"2004.10190","version":2},"attestation_state":"computed","paper":{"title":"Never Stop Learning: The Effectiveness of Fine-Tuning in Robotic Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Benjamin Swanson, Chelsea Finn, Gaurav S. Sukhatme, Karol Hausman, Ryan Julian, Sergey Levine","submitted_at":"2020-04-21T17:57:04Z","abstract_excerpt":"One of the great promises of robot learning systems is that they will be able to learn from their mistakes and continuously adapt to ever-changing environments. Despite this potential, most of the robot learning systems today are deployed as a fixed policy and they are not being adapted after their deployment. Can we efficiently adapt previously learned behaviors to new environments, objects and percepts in the real world? In this paper, we present a method and empirical evidence towards a robot learning framework that facilitates continuous adaption. In particular, we demonstrate how to adapt"},"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":"2004.10190","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-04-21T17:57:04Z","cross_cats_sorted":["cs.CV","cs.RO","stat.ML"],"title_canon_sha256":"5c8b08fc52ba500b9b2259a3d3c12de1bf0a13a23164b64cd80482f14360b32e","abstract_canon_sha256":"e05c63252114763e4c83472ec5332f20d94587465023e2f2f862210fac877f29"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:23:46.850156Z","signature_b64":"R/HvoZ/bR9stwgWZLdHEVdj3WD7LSuv6mE7HaKn6ftMB/dv9TqWRoNSm9WBXSTjbI+ZWCDJpWigrNTWC9MNfAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c71b5f5367fe9ac8f635ee5c011c146f98cac13e4ba2ce6ef7dcbc1ab4b3760a","last_reissued_at":"2026-07-05T01:23:46.849691Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:23:46.849691Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Never Stop Learning: The Effectiveness of Fine-Tuning in Robotic Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Benjamin Swanson, Chelsea Finn, Gaurav S. Sukhatme, Karol Hausman, Ryan Julian, Sergey Levine","submitted_at":"2020-04-21T17:57:04Z","abstract_excerpt":"One of the great promises of robot learning systems is that they will be able to learn from their mistakes and continuously adapt to ever-changing environments. Despite this potential, most of the robot learning systems today are deployed as a fixed policy and they are not being adapted after their deployment. Can we efficiently adapt previously learned behaviors to new environments, objects and percepts in the real world? In this paper, we present a method and empirical evidence towards a robot learning framework that facilitates continuous adaption. In particular, we demonstrate how to adapt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.10190","kind":"arxiv","version":2},"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/2004.10190/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":"2004.10190","created_at":"2026-07-05T01:23:46.849741+00:00"},{"alias_kind":"arxiv_version","alias_value":"2004.10190v2","created_at":"2026-07-05T01:23:46.849741+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.10190","created_at":"2026-07-05T01:23:46.849741+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y4NV6U3H72NM","created_at":"2026-07-05T01:23:46.849741+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y4NV6U3H72NMR5RV","created_at":"2026-07-05T01:23:46.849741+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y4NV6U3H","created_at":"2026-07-05T01:23:46.849741+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22145","citing_title":"Zero-shot Transfer of Reinforcement Learning Control Policies for the Swing-Up and Stabilization of a Cart-Pole System","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2507.09180","citing_title":"Multimodal Fusion for Sim2real Transfer in Visual Reinforcement Learning","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2505.18719","citing_title":"VLA-RL: Towards Masterful and General Robotic Manipulation with Scalable Reinforcement Learning","ref_index":34,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6","json":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6.json","graph_json":"https://pith.science/api/pith-number/Y4NV6U3H72NMR5RV5ZOACHAUN6/graph.json","events_json":"https://pith.science/api/pith-number/Y4NV6U3H72NMR5RV5ZOACHAUN6/events.json","paper":"https://pith.science/paper/Y4NV6U3H"},"agent_actions":{"view_html":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6","download_json":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6.json","view_paper":"https://pith.science/paper/Y4NV6U3H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2004.10190&json=true","fetch_graph":"https://pith.science/api/pith-number/Y4NV6U3H72NMR5RV5ZOACHAUN6/graph.json","fetch_events":"https://pith.science/api/pith-number/Y4NV6U3H72NMR5RV5ZOACHAUN6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6/action/storage_attestation","attest_author":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6/action/author_attestation","sign_citation":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6/action/citation_signature","submit_replication":"https://pith.science/pith/Y4NV6U3H72NMR5RV5ZOACHAUN6/action/replication_record"}},"created_at":"2026-07-05T01:23:46.849741+00:00","updated_at":"2026-07-05T01:23:46.849741+00:00"}