{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:OTL5L3CTH4MWNVZYIKTE7UA7NI","short_pith_number":"pith:OTL5L3CT","schema_version":"1.0","canonical_sha256":"74d7d5ec533f1966d73842a64fd01f6a3e135063379e89b4e18fe49f9ff91f6c","source":{"kind":"arxiv","id":"2209.13900","version":1},"attestation_state":"computed","paper":{"title":"Disentangling Transfer in Continual Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"{\\L}ukasz Kuci\\'nski, Maciej Wo{\\l}czyk, Micha{\\l} Zaj\\k{a}c, Piotr Mi{\\l}o\\'s, Razvan Pascanu","submitted_at":"2022-09-28T08:01:09Z","abstract_excerpt":"The ability of continual learning systems to transfer knowledge from previously seen tasks in order to maximize performance on new tasks is a significant challenge for the field, limiting the applicability of continual learning solutions to realistic scenarios. Consequently, this study aims to broaden our understanding of transfer and its driving forces in the specific case of continual reinforcement learning. We adopt SAC as the underlying RL algorithm and Continual World as a suite of continuous control tasks. We systematically study how different components of SAC (the actor and the critic,"},"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":"2209.13900","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-09-28T08:01:09Z","cross_cats_sorted":[],"title_canon_sha256":"05828b54c78a3e3cc827c8175c39b671a16c0a27d7889174a856a6133fca9808","abstract_canon_sha256":"96dd36414106401400fa2f8f4ee621596d51241d5856cdd1123e30212e283a29"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:01:45.365809Z","signature_b64":"7IDIaGT+j8rIY5uwoocXIr0ZTW4EbeiFdKBZK2x0AiyOYex51Hzt4N2lbLmP70nlUjhswoVNLh8TFiXHwsdnDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"74d7d5ec533f1966d73842a64fd01f6a3e135063379e89b4e18fe49f9ff91f6c","last_reissued_at":"2026-07-05T05:01:45.365379Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:01:45.365379Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Disentangling Transfer in Continual Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"{\\L}ukasz Kuci\\'nski, Maciej Wo{\\l}czyk, Micha{\\l} Zaj\\k{a}c, Piotr Mi{\\l}o\\'s, Razvan Pascanu","submitted_at":"2022-09-28T08:01:09Z","abstract_excerpt":"The ability of continual learning systems to transfer knowledge from previously seen tasks in order to maximize performance on new tasks is a significant challenge for the field, limiting the applicability of continual learning solutions to realistic scenarios. Consequently, this study aims to broaden our understanding of transfer and its driving forces in the specific case of continual reinforcement learning. We adopt SAC as the underlying RL algorithm and Continual World as a suite of continuous control tasks. We systematically study how different components of SAC (the actor and the critic,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.13900","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/2209.13900/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":"2209.13900","created_at":"2026-07-05T05:01:45.365436+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.13900v1","created_at":"2026-07-05T05:01:45.365436+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.13900","created_at":"2026-07-05T05:01:45.365436+00:00"},{"alias_kind":"pith_short_12","alias_value":"OTL5L3CTH4MW","created_at":"2026-07-05T05:01:45.365436+00:00"},{"alias_kind":"pith_short_16","alias_value":"OTL5L3CTH4MWNVZY","created_at":"2026-07-05T05:01:45.365436+00:00"},{"alias_kind":"pith_short_8","alias_value":"OTL5L3CT","created_at":"2026-07-05T05:01:45.365436+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2306.03310","citing_title":"LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning","ref_index":70,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI","json":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI.json","graph_json":"https://pith.science/api/pith-number/OTL5L3CTH4MWNVZYIKTE7UA7NI/graph.json","events_json":"https://pith.science/api/pith-number/OTL5L3CTH4MWNVZYIKTE7UA7NI/events.json","paper":"https://pith.science/paper/OTL5L3CT"},"agent_actions":{"view_html":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI","download_json":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI.json","view_paper":"https://pith.science/paper/OTL5L3CT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.13900&json=true","fetch_graph":"https://pith.science/api/pith-number/OTL5L3CTH4MWNVZYIKTE7UA7NI/graph.json","fetch_events":"https://pith.science/api/pith-number/OTL5L3CTH4MWNVZYIKTE7UA7NI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI/action/storage_attestation","attest_author":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI/action/author_attestation","sign_citation":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI/action/citation_signature","submit_replication":"https://pith.science/pith/OTL5L3CTH4MWNVZYIKTE7UA7NI/action/replication_record"}},"created_at":"2026-07-05T05:01:45.365436+00:00","updated_at":"2026-07-05T05:01:45.365436+00:00"}