{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:42UOQNSESZXWXMZNSTZ7QMCKMW","short_pith_number":"pith:42UOQNSE","schema_version":"1.0","canonical_sha256":"e6a8e83644966f6bb32d94f3f8304a658d6a71949f46edf9419dd7bae9dcecdf","source":{"kind":"arxiv","id":"2403.06571","version":2},"attestation_state":"computed","paper":{"title":"Scalable Online Exploration via Coverability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Akshay Krishnamurthy, Dylan J. Foster, Philip Amortila","submitted_at":"2024-03-11T10:14:06Z","abstract_excerpt":"Exploration is a major challenge in reinforcement learning, especially for high-dimensional domains that require function approximation. We propose exploration objectives -- policy optimization objectives that enable downstream maximization of any reward function -- as a conceptual framework to systematize the study of exploration. Within this framework, we introduce a new objective, $L_1$-Coverage, which generalizes previous exploration schemes and supports three fundamental desiderata:\n  1. Intrinsic complexity control. $L_1$-Coverage is associated with a structural parameter, $L_1$-Coverabi"},"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":"2403.06571","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-03-11T10:14:06Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"75c95bf4a3f2be0d4026514700e3dd8cd6f7533f49d14487ed80560d56b725f6","abstract_canon_sha256":"477981a413742cd6382f46fb64180f26f40ccc54f2eba36403110733ec0849e5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:25.407378Z","signature_b64":"tscgOTDXbbNMMHUReJpD6PqFRtdrQVNRsCXP6E9crpdBhSSddJInzy1HXtRkZ4d9tof0wE7MXyZivq0IJ3ZTCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e6a8e83644966f6bb32d94f3f8304a658d6a71949f46edf9419dd7bae9dcecdf","last_reissued_at":"2026-07-05T08:27:25.406909Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:25.406909Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable Online Exploration via Coverability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Akshay Krishnamurthy, Dylan J. Foster, Philip Amortila","submitted_at":"2024-03-11T10:14:06Z","abstract_excerpt":"Exploration is a major challenge in reinforcement learning, especially for high-dimensional domains that require function approximation. We propose exploration objectives -- policy optimization objectives that enable downstream maximization of any reward function -- as a conceptual framework to systematize the study of exploration. Within this framework, we introduce a new objective, $L_1$-Coverage, which generalizes previous exploration schemes and supports three fundamental desiderata:\n  1. Intrinsic complexity control. $L_1$-Coverage is associated with a structural parameter, $L_1$-Coverabi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.06571","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2403.06571/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":"2403.06571","created_at":"2026-07-05T08:27:25.406974+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.06571v2","created_at":"2026-07-05T08:27:25.406974+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.06571","created_at":"2026-07-05T08:27:25.406974+00:00"},{"alias_kind":"pith_short_12","alias_value":"42UOQNSESZXW","created_at":"2026-07-05T08:27:25.406974+00:00"},{"alias_kind":"pith_short_16","alias_value":"42UOQNSESZXWXMZN","created_at":"2026-07-05T08:27:25.406974+00:00"},{"alias_kind":"pith_short_8","alias_value":"42UOQNSE","created_at":"2026-07-05T08:27:25.406974+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.21395","citing_title":"Square$\\chi$PO: Differentially Private and Robust $\\chi^2$-Preference Optimization in Offline Direct Alignment","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW","json":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW.json","graph_json":"https://pith.science/api/pith-number/42UOQNSESZXWXMZNSTZ7QMCKMW/graph.json","events_json":"https://pith.science/api/pith-number/42UOQNSESZXWXMZNSTZ7QMCKMW/events.json","paper":"https://pith.science/paper/42UOQNSE"},"agent_actions":{"view_html":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW","download_json":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW.json","view_paper":"https://pith.science/paper/42UOQNSE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.06571&json=true","fetch_graph":"https://pith.science/api/pith-number/42UOQNSESZXWXMZNSTZ7QMCKMW/graph.json","fetch_events":"https://pith.science/api/pith-number/42UOQNSESZXWXMZNSTZ7QMCKMW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW/action/storage_attestation","attest_author":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW/action/author_attestation","sign_citation":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW/action/citation_signature","submit_replication":"https://pith.science/pith/42UOQNSESZXWXMZNSTZ7QMCKMW/action/replication_record"}},"created_at":"2026-07-05T08:27:25.406974+00:00","updated_at":"2026-07-05T08:27:25.406974+00:00"}