{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:STAOABRLMVOYMJ5OBLH5EPHIZ6","short_pith_number":"pith:STAOABRL","schema_version":"1.0","canonical_sha256":"94c0e0062b655d8627ae0acfd23ce8cfaa43b9e21facae8f26d6072896aff4d1","source":{"kind":"arxiv","id":"2501.07761","version":2},"attestation_state":"computed","paper":{"title":"Impatient Bandits: Optimizing for the Long-Term Without Delay","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Russo, Kamil Ciosek, Kelly W. Zhang, Lucas Maystre, Thomas Baldwin-McDonald","submitted_at":"2025-01-14T00:28:26Z","abstract_excerpt":"Increasingly, recommender systems are tasked with improving users' long-term satisfaction. In this context, we study a content exploration task, which we formalize as a bandit problem with delayed rewards. There is an apparent trade-off in choosing the learning signal: waiting for the full reward to become available might take several weeks, slowing the rate of learning, whereas using short-term proxy rewards reflects the actual long-term goal only imperfectly. First, we develop a predictive model of delayed rewards that incorporates all information obtained to date. Rewards as well as shorter"},"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":"2501.07761","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-14T00:28:26Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"6e7a22fa07a37e8cadeba4a429736dbdc17e3246b0154635bf216eb28bfcbc5c","abstract_canon_sha256":"e1c249c287792ae70393029d8a7f780cc1ffff4658e35817bb2544d98a56e697"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T00:14:20.309554Z","signature_b64":"n7L9v4um9g0vt9ZtDKVhDt8wqdwAXO95plWhcZYKjG6A13kNiAk31OYaCZgbuwgNnPWjEiU8GFlxpMjwXzteBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"94c0e0062b655d8627ae0acfd23ce8cfaa43b9e21facae8f26d6072896aff4d1","last_reissued_at":"2026-06-24T00:14:20.309024Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T00:14:20.309024Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Impatient Bandits: Optimizing for the Long-Term Without Delay","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Russo, Kamil Ciosek, Kelly W. Zhang, Lucas Maystre, Thomas Baldwin-McDonald","submitted_at":"2025-01-14T00:28:26Z","abstract_excerpt":"Increasingly, recommender systems are tasked with improving users' long-term satisfaction. In this context, we study a content exploration task, which we formalize as a bandit problem with delayed rewards. There is an apparent trade-off in choosing the learning signal: waiting for the full reward to become available might take several weeks, slowing the rate of learning, whereas using short-term proxy rewards reflects the actual long-term goal only imperfectly. First, we develop a predictive model of delayed rewards that incorporates all information obtained to date. Rewards as well as shorter"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.07761","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/2501.07761/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":"2501.07761","created_at":"2026-06-24T00:14:20.309088+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.07761v2","created_at":"2026-06-24T00:14:20.309088+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.07761","created_at":"2026-06-24T00:14:20.309088+00:00"},{"alias_kind":"pith_short_12","alias_value":"STAOABRLMVOY","created_at":"2026-06-24T00:14:20.309088+00:00"},{"alias_kind":"pith_short_16","alias_value":"STAOABRLMVOYMJ5O","created_at":"2026-06-24T00:14:20.309088+00:00"},{"alias_kind":"pith_short_8","alias_value":"STAOABRL","created_at":"2026-06-24T00:14:20.309088+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2606.00913","citing_title":"Bandit Simulation for Average Reward Inference","ref_index":14,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6","json":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6.json","graph_json":"https://pith.science/api/pith-number/STAOABRLMVOYMJ5OBLH5EPHIZ6/graph.json","events_json":"https://pith.science/api/pith-number/STAOABRLMVOYMJ5OBLH5EPHIZ6/events.json","paper":"https://pith.science/paper/STAOABRL"},"agent_actions":{"view_html":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6","download_json":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6.json","view_paper":"https://pith.science/paper/STAOABRL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.07761&json=true","fetch_graph":"https://pith.science/api/pith-number/STAOABRLMVOYMJ5OBLH5EPHIZ6/graph.json","fetch_events":"https://pith.science/api/pith-number/STAOABRLMVOYMJ5OBLH5EPHIZ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6/action/storage_attestation","attest_author":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6/action/author_attestation","sign_citation":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6/action/citation_signature","submit_replication":"https://pith.science/pith/STAOABRLMVOYMJ5OBLH5EPHIZ6/action/replication_record"}},"created_at":"2026-06-24T00:14:20.309088+00:00","updated_at":"2026-06-24T00:14:20.309088+00:00"}