{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:4TF2PFST4AYJSP7V6CL26BVBGV","short_pith_number":"pith:4TF2PFST","schema_version":"1.0","canonical_sha256":"e4cba79653e030993ff5f097af06a1355f80f24e999550f80a45a405c9521d23","source":{"kind":"arxiv","id":"1704.03732","version":4},"attestation_state":"computed","paper":{"title":"Deep Q-learning from Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Andrew Sendonaris, Audrunas Gruslys, Bilal Piot, Dan Horgan, Gabriel Dulac-Arnold, Ian Osband, Joel Z. Leibo, John Agapiou, John Quan, Marc Lanctot, Matej Vecerik, Olivier Pietquin, Todd Hester, Tom Schaul","submitted_at":"2017-04-12T12:44:37Z","abstract_excerpt":"Deep reinforcement learning (RL) has achieved several high profile successes in difficult decision-making problems. However, these algorithms typically require a huge amount of data before they reach reasonable performance. In fact, their performance during learning can be extremely poor. This may be acceptable for a simulator, but it severely limits the applicability of deep RL to many real-world tasks, where the agent must learn in the real environment. In this paper we study a setting where the agent may access data from previous control of the system. We present an algorithm, Deep Q-learni"},"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":"1704.03732","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2017-04-12T12:44:37Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"15c6de0049e5300bd9e46aa894cc50b67c4421d0e641e7d747b89b398a32b2c5","abstract_canon_sha256":"be25f9b75ff99dea0eb1500a831034d786bd9e8897e0b1903705bf8cdeda07b5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:29:46.197389Z","signature_b64":"SSfsDGOhD1GQPV4Yx+ClJRP4GEHljiO/pJtvj2JBM8mQ1YrjhUr73+iPbjH81Bo2tLYajmZ7KwgWRwFDDlq3AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4cba79653e030993ff5f097af06a1355f80f24e999550f80a45a405c9521d23","last_reissued_at":"2026-05-18T00:29:46.196823Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:29:46.196823Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Q-learning from Demonstrations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Andrew Sendonaris, Audrunas Gruslys, Bilal Piot, Dan Horgan, Gabriel Dulac-Arnold, Ian Osband, Joel Z. Leibo, John Agapiou, John Quan, Marc Lanctot, Matej Vecerik, Olivier Pietquin, Todd Hester, Tom Schaul","submitted_at":"2017-04-12T12:44:37Z","abstract_excerpt":"Deep reinforcement learning (RL) has achieved several high profile successes in difficult decision-making problems. However, these algorithms typically require a huge amount of data before they reach reasonable performance. In fact, their performance during learning can be extremely poor. This may be acceptable for a simulator, but it severely limits the applicability of deep RL to many real-world tasks, where the agent must learn in the real environment. In this paper we study a setting where the agent may access data from previous control of the system. We present an algorithm, Deep Q-learni"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.03732","kind":"arxiv","version":4},"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":"1704.03732","created_at":"2026-05-18T00:29:46.196921+00:00"},{"alias_kind":"arxiv_version","alias_value":"1704.03732v4","created_at":"2026-05-18T00:29:46.196921+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.03732","created_at":"2026-05-18T00:29:46.196921+00:00"},{"alias_kind":"pith_short_12","alias_value":"4TF2PFST4AYJ","created_at":"2026-05-18T12:31:00.734936+00:00"},{"alias_kind":"pith_short_16","alias_value":"4TF2PFST4AYJSP7V","created_at":"2026-05-18T12:31:00.734936+00:00"},{"alias_kind":"pith_short_8","alias_value":"4TF2PFST","created_at":"2026-05-18T12:31:00.734936+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2507.07986","citing_title":"EXPO: Stable Reinforcement Learning with Expressive Policies","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"1706.03741","citing_title":"Deep reinforcement learning from human preferences","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV","json":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV.json","graph_json":"https://pith.science/api/pith-number/4TF2PFST4AYJSP7V6CL26BVBGV/graph.json","events_json":"https://pith.science/api/pith-number/4TF2PFST4AYJSP7V6CL26BVBGV/events.json","paper":"https://pith.science/paper/4TF2PFST"},"agent_actions":{"view_html":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV","download_json":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV.json","view_paper":"https://pith.science/paper/4TF2PFST","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1704.03732&json=true","fetch_graph":"https://pith.science/api/pith-number/4TF2PFST4AYJSP7V6CL26BVBGV/graph.json","fetch_events":"https://pith.science/api/pith-number/4TF2PFST4AYJSP7V6CL26BVBGV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV/action/storage_attestation","attest_author":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV/action/author_attestation","sign_citation":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV/action/citation_signature","submit_replication":"https://pith.science/pith/4TF2PFST4AYJSP7V6CL26BVBGV/action/replication_record"}},"created_at":"2026-05-18T00:29:46.196921+00:00","updated_at":"2026-05-18T00:29:46.196921+00:00"}