{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:LGYWERFX2B2LW2OOGPWSYMG4LS","short_pith_number":"pith:LGYWERFX","schema_version":"1.0","canonical_sha256":"59b16244b7d074bb69ce33ed2c30dc5cbeb9b3e3e164ca5d70568ed023bba1fe","source":{"kind":"arxiv","id":"1901.10964","version":1},"attestation_state":"computed","paper":{"title":"Transfer in Deep Reinforcement Learning Using Successor Features and Generalised Policy Improvement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andr\\'e Barreto, Augustin \\v{Z}\\'idek, Daniel Mankowitz, David Silver, Diana Borsa, John Quan, Matteo Hessel, R\\'emi Munos, Tom Schaul","submitted_at":"2019-01-30T17:30:31Z","abstract_excerpt":"The ability to transfer skills across tasks has the potential to scale up reinforcement learning (RL) agents to environments currently out of reach. Recently, a framework based on two ideas, successor features (SFs) and generalised policy improvement (GPI), has been introduced as a principled way of transferring skills. In this paper we extend the SFs & GPI framework in two ways. One of the basic assumptions underlying the original formulation of SFs & GPI is that rewards for all tasks of interest can be computed as linear combinations of a fixed set of features. We relax this constraint and s"},"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":"1901.10964","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-30T17:30:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a792c9919db3b16455c6c1e84c332c360c0b50822efad2f0cd340fbccd78d7a0","abstract_canon_sha256":"30c3e40d2b61e9665ad670117adad615e0e96baba687981420a67349f3d472fe"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:55:05.665436Z","signature_b64":"cC/nB+ZGsLgn3BtRfYtFBXzakLh5oMBYtGWQhz1JiFWttqko1i2X/lA66B+OTTe5UMTIGh3A21sb+CCNyFZ+Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"59b16244b7d074bb69ce33ed2c30dc5cbeb9b3e3e164ca5d70568ed023bba1fe","last_reissued_at":"2026-05-17T23:55:05.664974Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:55:05.664974Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Transfer in Deep Reinforcement Learning Using Successor Features and Generalised Policy Improvement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Andr\\'e Barreto, Augustin \\v{Z}\\'idek, Daniel Mankowitz, David Silver, Diana Borsa, John Quan, Matteo Hessel, R\\'emi Munos, Tom Schaul","submitted_at":"2019-01-30T17:30:31Z","abstract_excerpt":"The ability to transfer skills across tasks has the potential to scale up reinforcement learning (RL) agents to environments currently out of reach. Recently, a framework based on two ideas, successor features (SFs) and generalised policy improvement (GPI), has been introduced as a principled way of transferring skills. In this paper we extend the SFs & GPI framework in two ways. One of the basic assumptions underlying the original formulation of SFs & GPI is that rewards for all tasks of interest can be computed as linear combinations of a fixed set of features. We relax this constraint and s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.10964","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":""},"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":"1901.10964","created_at":"2026-05-17T23:55:05.665055+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.10964v1","created_at":"2026-05-17T23:55:05.665055+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.10964","created_at":"2026-05-17T23:55:05.665055+00:00"},{"alias_kind":"pith_short_12","alias_value":"LGYWERFX2B2L","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_16","alias_value":"LGYWERFX2B2LW2OO","created_at":"2026-05-18T12:33:21.387695+00:00"},{"alias_kind":"pith_short_8","alias_value":"LGYWERFX","created_at":"2026-05-18T12:33:21.387695+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1906.09480","citing_title":"A neurally plausible model learns successor representations in partially observable environments","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2604.20174","citing_title":"Lever: Inference-Time Policy Reuse under Support Constraints","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS","json":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS.json","graph_json":"https://pith.science/api/pith-number/LGYWERFX2B2LW2OOGPWSYMG4LS/graph.json","events_json":"https://pith.science/api/pith-number/LGYWERFX2B2LW2OOGPWSYMG4LS/events.json","paper":"https://pith.science/paper/LGYWERFX"},"agent_actions":{"view_html":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS","download_json":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS.json","view_paper":"https://pith.science/paper/LGYWERFX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.10964&json=true","fetch_graph":"https://pith.science/api/pith-number/LGYWERFX2B2LW2OOGPWSYMG4LS/graph.json","fetch_events":"https://pith.science/api/pith-number/LGYWERFX2B2LW2OOGPWSYMG4LS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS/action/storage_attestation","attest_author":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS/action/author_attestation","sign_citation":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS/action/citation_signature","submit_replication":"https://pith.science/pith/LGYWERFX2B2LW2OOGPWSYMG4LS/action/replication_record"}},"created_at":"2026-05-17T23:55:05.665055+00:00","updated_at":"2026-05-17T23:55:05.665055+00:00"}