{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:F6NNLIBXX2ON36U5T6NRNLZYIY","short_pith_number":"pith:F6NNLIBX","schema_version":"1.0","canonical_sha256":"2f9ad5a037be9cddfa9d9f9b16af38460095ec7224c0f4402e657e263a0a1f20","source":{"kind":"arxiv","id":"2306.02865","version":5},"attestation_state":"computed","paper":{"title":"Seizing Serendipity: Exploiting the Value of Past Success in Off-Policy Actor-Critic","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fuchun Sun, Huazhe Xu, Jianwei Zhang, Tianying Ji, Xianyuan Zhan, Yu Luo","submitted_at":"2023-06-05T13:38:14Z","abstract_excerpt":"Learning high-quality $Q$-value functions plays a key role in the success of many modern off-policy deep reinforcement learning (RL) algorithms. Previous works primarily focus on addressing the value overestimation issue, an outcome of adopting function approximators and off-policy learning. Deviating from the common viewpoint, we observe that $Q$-values are often underestimated in the latter stage of the RL training process, potentially hindering policy learning and reducing sample efficiency. We find that such a long-neglected phenomenon is often related to the use of inferior actions from t"},"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":"2306.02865","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-05T13:38:14Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"bf8422c8538f736a1eebc9ea263337e1b1841fce0c14151ac2d610e1c42707f9","abstract_canon_sha256":"f6603f0d72e364f79669178e005da88f60b950bf64ebe8a7426bf55404f2c1c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:17:57.289052Z","signature_b64":"dEblSJ+xbz2kFptGBdrSBSalZhnmsu74vNEtzHPUlpJxNokLI+DqYJ5tplVPhgeyrH6nzk7HJzbbR5xiL7LSCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f9ad5a037be9cddfa9d9f9b16af38460095ec7224c0f4402e657e263a0a1f20","last_reissued_at":"2026-07-05T08:17:57.288558Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:17:57.288558Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Seizing Serendipity: Exploiting the Value of Past Success in Off-Policy Actor-Critic","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Fuchun Sun, Huazhe Xu, Jianwei Zhang, Tianying Ji, Xianyuan Zhan, Yu Luo","submitted_at":"2023-06-05T13:38:14Z","abstract_excerpt":"Learning high-quality $Q$-value functions plays a key role in the success of many modern off-policy deep reinforcement learning (RL) algorithms. Previous works primarily focus on addressing the value overestimation issue, an outcome of adopting function approximators and off-policy learning. Deviating from the common viewpoint, we observe that $Q$-values are often underestimated in the latter stage of the RL training process, potentially hindering policy learning and reducing sample efficiency. We find that such a long-neglected phenomenon is often related to the use of inferior actions from t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.02865","kind":"arxiv","version":5},"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/2306.02865/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":"2306.02865","created_at":"2026-07-05T08:17:57.288612+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.02865v5","created_at":"2026-07-05T08:17:57.288612+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.02865","created_at":"2026-07-05T08:17:57.288612+00:00"},{"alias_kind":"pith_short_12","alias_value":"F6NNLIBXX2ON","created_at":"2026-07-05T08:17:57.288612+00:00"},{"alias_kind":"pith_short_16","alias_value":"F6NNLIBXX2ON36U5","created_at":"2026-07-05T08:17:57.288612+00:00"},{"alias_kind":"pith_short_8","alias_value":"F6NNLIBX","created_at":"2026-07-05T08:17:57.288612+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.07762","citing_title":"Efficient Online Reinforcement Learning Fine-Tuning Need Not Retain Offline Data","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY","json":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY.json","graph_json":"https://pith.science/api/pith-number/F6NNLIBXX2ON36U5T6NRNLZYIY/graph.json","events_json":"https://pith.science/api/pith-number/F6NNLIBXX2ON36U5T6NRNLZYIY/events.json","paper":"https://pith.science/paper/F6NNLIBX"},"agent_actions":{"view_html":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY","download_json":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY.json","view_paper":"https://pith.science/paper/F6NNLIBX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.02865&json=true","fetch_graph":"https://pith.science/api/pith-number/F6NNLIBXX2ON36U5T6NRNLZYIY/graph.json","fetch_events":"https://pith.science/api/pith-number/F6NNLIBXX2ON36U5T6NRNLZYIY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY/action/storage_attestation","attest_author":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY/action/author_attestation","sign_citation":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY/action/citation_signature","submit_replication":"https://pith.science/pith/F6NNLIBXX2ON36U5T6NRNLZYIY/action/replication_record"}},"created_at":"2026-07-05T08:17:57.288612+00:00","updated_at":"2026-07-05T08:17:57.288612+00:00"}