{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:LU67JPGQBQE4MLYKLRJPNPGIFT","short_pith_number":"pith:LU67JPGQ","schema_version":"1.0","canonical_sha256":"5d3df4bcd00c09c62f0a5c52f6bcc82cdc09c99d57d1ba2111d512dec755f1e7","source":{"kind":"arxiv","id":"2011.07931","version":1},"attestation_state":"computed","paper":{"title":"Do Offline Metrics Predict Online Performance in Recommender Systems?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Alex Zhao, Benjamin Recht, Karl Krauth, Michael I. Jordan, Mihaela Curmei, Sarah Dean, Wenshuo Guo","submitted_at":"2020-11-07T01:41:13Z","abstract_excerpt":"Recommender systems operate in an inherently dynamical setting. Past recommendations influence future behavior, including which data points are observed and how user preferences change. However, experimenting in production systems with real user dynamics is often infeasible, and existing simulation-based approaches have limited scale. As a result, many state-of-the-art algorithms are designed to solve supervised learning problems, and progress is judged only by offline metrics. In this work we investigate the extent to which offline metrics predict online performance by evaluating eleven recom"},"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":"2011.07931","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.IR","submitted_at":"2020-11-07T01:41:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3878bb2e071f6b6cf9fd7ef0c515e6d93d4f02668e04ce26ede26cf352cd6014","abstract_canon_sha256":"17bd677dc7958d1ccbbe6149038925f7d00531a3ceb2803f5b955515abcea35a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:51:55.742822Z","signature_b64":"HWlV3ETrciV2PHhMGlFt2bUjbdzu/btpQnG8rGFb8mogHXULOXDlCOhMHLvlCfRj7KeWdgWVvlgiY/lQVTFrDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5d3df4bcd00c09c62f0a5c52f6bcc82cdc09c99d57d1ba2111d512dec755f1e7","last_reissued_at":"2026-07-05T01:51:55.742471Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:51:55.742471Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Offline Metrics Predict Online Performance in Recommender Systems?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.IR","authors_text":"Alex Zhao, Benjamin Recht, Karl Krauth, Michael I. Jordan, Mihaela Curmei, Sarah Dean, Wenshuo Guo","submitted_at":"2020-11-07T01:41:13Z","abstract_excerpt":"Recommender systems operate in an inherently dynamical setting. Past recommendations influence future behavior, including which data points are observed and how user preferences change. However, experimenting in production systems with real user dynamics is often infeasible, and existing simulation-based approaches have limited scale. As a result, many state-of-the-art algorithms are designed to solve supervised learning problems, and progress is judged only by offline metrics. In this work we investigate the extent to which offline metrics predict online performance by evaluating eleven recom"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.07931","kind":"arxiv","version":1},"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/2011.07931/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":"2011.07931","created_at":"2026-07-05T01:51:55.742525+00:00"},{"alias_kind":"arxiv_version","alias_value":"2011.07931v1","created_at":"2026-07-05T01:51:55.742525+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.07931","created_at":"2026-07-05T01:51:55.742525+00:00"},{"alias_kind":"pith_short_12","alias_value":"LU67JPGQBQE4","created_at":"2026-07-05T01:51:55.742525+00:00"},{"alias_kind":"pith_short_16","alias_value":"LU67JPGQBQE4MLYK","created_at":"2026-07-05T01:51:55.742525+00:00"},{"alias_kind":"pith_short_8","alias_value":"LU67JPGQ","created_at":"2026-07-05T01:51:55.742525+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.21750","citing_title":"Multistakeholder Impacts of Profile Portability in a Recommender Ecosystem","ref_index":44,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT","json":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT.json","graph_json":"https://pith.science/api/pith-number/LU67JPGQBQE4MLYKLRJPNPGIFT/graph.json","events_json":"https://pith.science/api/pith-number/LU67JPGQBQE4MLYKLRJPNPGIFT/events.json","paper":"https://pith.science/paper/LU67JPGQ"},"agent_actions":{"view_html":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT","download_json":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT.json","view_paper":"https://pith.science/paper/LU67JPGQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2011.07931&json=true","fetch_graph":"https://pith.science/api/pith-number/LU67JPGQBQE4MLYKLRJPNPGIFT/graph.json","fetch_events":"https://pith.science/api/pith-number/LU67JPGQBQE4MLYKLRJPNPGIFT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT/action/storage_attestation","attest_author":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT/action/author_attestation","sign_citation":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT/action/citation_signature","submit_replication":"https://pith.science/pith/LU67JPGQBQE4MLYKLRJPNPGIFT/action/replication_record"}},"created_at":"2026-07-05T01:51:55.742525+00:00","updated_at":"2026-07-05T01:51:55.742525+00:00"}