{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:Y2WI6TZMOB5LOBPTBGYT6UGTTI","short_pith_number":"pith:Y2WI6TZM","schema_version":"1.0","canonical_sha256":"c6ac8f4f2c707ab705f309b13f50d39a291cf9320d4ec65bb84b0f05fcfe87d5","source":{"kind":"arxiv","id":"1706.04208","version":2},"attestation_state":"computed","paper":{"title":"Hybrid Reward Architecture for Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Harm van Seijen, Jeffrey Tsang, Joshua Romoff, Mehdi Fatemi, Romain Laroche, Tavian Barnes","submitted_at":"2017-06-13T18:05:48Z","abstract_excerpt":"One of the main challenges in reinforcement learning (RL) is generalisation. In typical deep RL methods this is achieved by approximating the optimal value function with a low-dimensional representation using a deep network. While this approach works well in many domains, in domains where the optimal value function cannot easily be reduced to a low-dimensional representation, learning can be very slow and unstable. This paper contributes towards tackling such challenging domains, by proposing a new method, called Hybrid Reward Architecture (HRA). HRA takes as input a decomposed reward function"},"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":"1706.04208","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-06-13T18:05:48Z","cross_cats_sorted":[],"title_canon_sha256":"78597ba42370b3065a53f37b48bbc43aee37c5ad4f3fa9e620fb446a25cdd248","abstract_canon_sha256":"2eebe47f992105fe4c0ea7b0c16d4e4967fac7077e8da510ea69bc87c65e8694"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:29:30.330073Z","signature_b64":"CvQM+zsC/2Ji4Ra5dx/I6mTwokVlyH3S+AeRf+GhfnoPSnXbXlSd8qTcExL5PuUf4bFgL1YgXE0Px+FTlf2yDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c6ac8f4f2c707ab705f309b13f50d39a291cf9320d4ec65bb84b0f05fcfe87d5","last_reissued_at":"2026-05-18T00:29:30.329426Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:29:30.329426Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hybrid Reward Architecture for Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Harm van Seijen, Jeffrey Tsang, Joshua Romoff, Mehdi Fatemi, Romain Laroche, Tavian Barnes","submitted_at":"2017-06-13T18:05:48Z","abstract_excerpt":"One of the main challenges in reinforcement learning (RL) is generalisation. In typical deep RL methods this is achieved by approximating the optimal value function with a low-dimensional representation using a deep network. While this approach works well in many domains, in domains where the optimal value function cannot easily be reduced to a low-dimensional representation, learning can be very slow and unstable. This paper contributes towards tackling such challenging domains, by proposing a new method, called Hybrid Reward Architecture (HRA). HRA takes as input a decomposed reward function"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1706.04208","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":""},"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":"1706.04208","created_at":"2026-05-18T00:29:30.329513+00:00"},{"alias_kind":"arxiv_version","alias_value":"1706.04208v2","created_at":"2026-05-18T00:29:30.329513+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1706.04208","created_at":"2026-05-18T00:29:30.329513+00:00"},{"alias_kind":"pith_short_12","alias_value":"Y2WI6TZMOB5L","created_at":"2026-05-18T12:31:56.362134+00:00"},{"alias_kind":"pith_short_16","alias_value":"Y2WI6TZMOB5LOBPT","created_at":"2026-05-18T12:31:56.362134+00:00"},{"alias_kind":"pith_short_8","alias_value":"Y2WI6TZM","created_at":"2026-05-18T12:31:56.362134+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/Y2WI6TZMOB5LOBPTBGYT6UGTTI","json":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI.json","graph_json":"https://pith.science/api/pith-number/Y2WI6TZMOB5LOBPTBGYT6UGTTI/graph.json","events_json":"https://pith.science/api/pith-number/Y2WI6TZMOB5LOBPTBGYT6UGTTI/events.json","paper":"https://pith.science/paper/Y2WI6TZM"},"agent_actions":{"view_html":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI","download_json":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI.json","view_paper":"https://pith.science/paper/Y2WI6TZM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1706.04208&json=true","fetch_graph":"https://pith.science/api/pith-number/Y2WI6TZMOB5LOBPTBGYT6UGTTI/graph.json","fetch_events":"https://pith.science/api/pith-number/Y2WI6TZMOB5LOBPTBGYT6UGTTI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI/action/storage_attestation","attest_author":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI/action/author_attestation","sign_citation":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI/action/citation_signature","submit_replication":"https://pith.science/pith/Y2WI6TZMOB5LOBPTBGYT6UGTTI/action/replication_record"}},"created_at":"2026-05-18T00:29:30.329513+00:00","updated_at":"2026-05-18T00:29:30.329513+00:00"}