{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:URTHULKXL2FGLDQTHWJMBNGAZI","short_pith_number":"pith:URTHULKX","schema_version":"1.0","canonical_sha256":"a4667a2d575e8a658e133d92c0b4c0ca34d1370b6122b59d903850e3b38820b3","source":{"kind":"arxiv","id":"2001.05411","version":3},"attestation_state":"computed","paper":{"title":"Lipschitz Lifelong Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Abel, Emmanuel Rachelson, Erwan Lecarpentier, Kavosh Asadi, Michael L. Littman, Yuu Jinnai","submitted_at":"2020-01-15T16:29:30Z","abstract_excerpt":"We consider the problem of knowledge transfer when an agent is facing a series of Reinforcement Learning (RL) tasks. We introduce a novel metric between Markov Decision Processes (MDPs) and establish that close MDPs have close optimal value functions. Formally, the optimal value functions are Lipschitz continuous with respect to the tasks space. These theoretical results lead us to a value-transfer method for Lifelong RL, which we use to build a PAC-MDP algorithm with improved convergence rate. Further, we show the method to experience no negative transfer with high probability. We illustrate "},"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":"2001.05411","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-01-15T16:29:30Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"0c699763278e595d5ecf6dc374c2d4cf94638a669971311095bda7028131a298","abstract_canon_sha256":"8072bd88e3350479b29646b37b67316026c52a05a3a769c902f6f9c06a0ca4e3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:25:09.728014Z","signature_b64":"c4iWm+Yb1ZWIwj5DrT4lX4q5m3V0B8ySfwOhX+yKESiPqOi9eRZX0/yo/tvLT4OUe41WIgzBKDeMAZIQSfOICw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a4667a2d575e8a658e133d92c0b4c0ca34d1370b6122b59d903850e3b38820b3","last_reissued_at":"2026-07-05T02:25:09.727563Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:25:09.727563Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lipschitz Lifelong Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Abel, Emmanuel Rachelson, Erwan Lecarpentier, Kavosh Asadi, Michael L. Littman, Yuu Jinnai","submitted_at":"2020-01-15T16:29:30Z","abstract_excerpt":"We consider the problem of knowledge transfer when an agent is facing a series of Reinforcement Learning (RL) tasks. We introduce a novel metric between Markov Decision Processes (MDPs) and establish that close MDPs have close optimal value functions. Formally, the optimal value functions are Lipschitz continuous with respect to the tasks space. These theoretical results lead us to a value-transfer method for Lifelong RL, which we use to build a PAC-MDP algorithm with improved convergence rate. Further, we show the method to experience no negative transfer with high probability. We illustrate "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2001.05411","kind":"arxiv","version":3},"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/2001.05411/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":"2001.05411","created_at":"2026-07-05T02:25:09.727621+00:00"},{"alias_kind":"arxiv_version","alias_value":"2001.05411v3","created_at":"2026-07-05T02:25:09.727621+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2001.05411","created_at":"2026-07-05T02:25:09.727621+00:00"},{"alias_kind":"pith_short_12","alias_value":"URTHULKXL2FG","created_at":"2026-07-05T02:25:09.727621+00:00"},{"alias_kind":"pith_short_16","alias_value":"URTHULKXL2FGLDQT","created_at":"2026-07-05T02:25:09.727621+00:00"},{"alias_kind":"pith_short_8","alias_value":"URTHULKX","created_at":"2026-07-05T02:25:09.727621+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00633","citing_title":"Lipschitz Lifelong Monte Carlo Tree Search for Mastering Non-Stationary Tasks","ref_index":2003,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI","json":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI.json","graph_json":"https://pith.science/api/pith-number/URTHULKXL2FGLDQTHWJMBNGAZI/graph.json","events_json":"https://pith.science/api/pith-number/URTHULKXL2FGLDQTHWJMBNGAZI/events.json","paper":"https://pith.science/paper/URTHULKX"},"agent_actions":{"view_html":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI","download_json":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI.json","view_paper":"https://pith.science/paper/URTHULKX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2001.05411&json=true","fetch_graph":"https://pith.science/api/pith-number/URTHULKXL2FGLDQTHWJMBNGAZI/graph.json","fetch_events":"https://pith.science/api/pith-number/URTHULKXL2FGLDQTHWJMBNGAZI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI/action/storage_attestation","attest_author":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI/action/author_attestation","sign_citation":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI/action/citation_signature","submit_replication":"https://pith.science/pith/URTHULKXL2FGLDQTHWJMBNGAZI/action/replication_record"}},"created_at":"2026-07-05T02:25:09.727621+00:00","updated_at":"2026-07-05T02:25:09.727621+00:00"}