{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:MOYWHB2HFBW5XDHO7EL3KZGI4X","short_pith_number":"pith:MOYWHB2H","schema_version":"1.0","canonical_sha256":"63b1638747286ddb8ceef917b564c8e5e2ef6a8d81df9278f2189d6550cb3e20","source":{"kind":"arxiv","id":"1909.12397","version":3},"attestation_state":"computed","paper":{"title":"CAQL: Continuous Action Q-Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Christian Tjandraatmadja, Craig Boutilier, Moonkyung Ryu, Ross Anderson, Yinlam Chow","submitted_at":"2019-09-26T21:16:17Z","abstract_excerpt":"Value-based reinforcement learning (RL) methods like Q-learning have shown success in a variety of domains. One challenge in applying Q-learning to continuous-action RL problems, however, is the continuous action maximization (max-Q) required for optimal Bellman backup. In this work, we develop CAQL, a (class of) algorithm(s) for continuous-action Q-learning that can use several plug-and-play optimizers for the max-Q problem. Leveraging recent optimization results for deep neural networks, we show that max-Q can be solved optimally using mixed-integer programming (MIP). When the Q-function rep"},"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":"1909.12397","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-26T21:16:17Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"7f7709d67010d7a5668bf43dcc4aa0fa0efa7635bbc571e26b1df3bd57a10fef","abstract_canon_sha256":"71343c2e20f32e4f22400cbc4344c23b5f65536ba3dc90c68c0b1a22bcce3cf9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:44:40.399791Z","signature_b64":"yUoH+9fqCYje3ezDcKBDkkGFDy/CZ93ctwVHb+Vge4fu2hqUnHKNYvIAMdP7SyxKhukQixuzKH5QiV/aaAEaCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"63b1638747286ddb8ceef917b564c8e5e2ef6a8d81df9278f2189d6550cb3e20","last_reissued_at":"2026-07-05T00:44:40.399410Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:44:40.399410Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CAQL: Continuous Action Q-Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Christian Tjandraatmadja, Craig Boutilier, Moonkyung Ryu, Ross Anderson, Yinlam Chow","submitted_at":"2019-09-26T21:16:17Z","abstract_excerpt":"Value-based reinforcement learning (RL) methods like Q-learning have shown success in a variety of domains. One challenge in applying Q-learning to continuous-action RL problems, however, is the continuous action maximization (max-Q) required for optimal Bellman backup. In this work, we develop CAQL, a (class of) algorithm(s) for continuous-action Q-learning that can use several plug-and-play optimizers for the max-Q problem. Leveraging recent optimization results for deep neural networks, we show that max-Q can be solved optimally using mixed-integer programming (MIP). When the Q-function rep"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.12397","kind":"arxiv","version":3},"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/1909.12397/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":"1909.12397","created_at":"2026-07-05T00:44:40.399466+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.12397v3","created_at":"2026-07-05T00:44:40.399466+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.12397","created_at":"2026-07-05T00:44:40.399466+00:00"},{"alias_kind":"pith_short_12","alias_value":"MOYWHB2HFBW5","created_at":"2026-07-05T00:44:40.399466+00:00"},{"alias_kind":"pith_short_16","alias_value":"MOYWHB2HFBW5XDHO","created_at":"2026-07-05T00:44:40.399466+00:00"},{"alias_kind":"pith_short_8","alias_value":"MOYWHB2H","created_at":"2026-07-05T00:44:40.399466+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.22746","citing_title":"Relaxation-Informed Training of Neural Network Surrogate Models","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X","json":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X.json","graph_json":"https://pith.science/api/pith-number/MOYWHB2HFBW5XDHO7EL3KZGI4X/graph.json","events_json":"https://pith.science/api/pith-number/MOYWHB2HFBW5XDHO7EL3KZGI4X/events.json","paper":"https://pith.science/paper/MOYWHB2H"},"agent_actions":{"view_html":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X","download_json":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X.json","view_paper":"https://pith.science/paper/MOYWHB2H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.12397&json=true","fetch_graph":"https://pith.science/api/pith-number/MOYWHB2HFBW5XDHO7EL3KZGI4X/graph.json","fetch_events":"https://pith.science/api/pith-number/MOYWHB2HFBW5XDHO7EL3KZGI4X/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X/action/storage_attestation","attest_author":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X/action/author_attestation","sign_citation":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X/action/citation_signature","submit_replication":"https://pith.science/pith/MOYWHB2HFBW5XDHO7EL3KZGI4X/action/replication_record"}},"created_at":"2026-07-05T00:44:40.399466+00:00","updated_at":"2026-07-05T00:44:40.399466+00:00"}