{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:ICT7BYMH3ZV5XZXAV5RNKOWQ4S","short_pith_number":"pith:ICT7BYMH","schema_version":"1.0","canonical_sha256":"40a7f0e187de6bdbe6e0af62d53ad0e4a69f66cbce1288b229d63e88c01cc0b6","source":{"kind":"arxiv","id":"2205.06811","version":2},"attestation_state":"computed","paper":{"title":"Nearly Optimal Algorithms for Linear Contextual Bandits with Adversarial Corruptions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dongruo Zhou, Jiafan He, Quanquan Gu, Tong Zhang","submitted_at":"2022-05-13T17:58:58Z","abstract_excerpt":"We study the linear contextual bandit problem in the presence of adversarial corruption, where the reward at each round is corrupted by an adversary, and the corruption level (i.e., the sum of corruption magnitudes over the horizon) is $C\\geq 0$. The best-known algorithms in this setting are limited in that they either are computationally inefficient or require a strong assumption on the corruption, or their regret is at least $C$ times worse than the regret without corruption. In this paper, to overcome these limitations, we propose a new algorithm based on the principle of optimism in the fa"},"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":"2205.06811","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-13T17:58:58Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"de93e4fb0c3f7e66148b0a4b8f612342a0f72a78b336e26ac727d4bf8f337651","abstract_canon_sha256":"367d468c5f37bf2fac7fa4b6c54525f6c457f3d1439c1e01c298d00014e27cf7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:38:50.627745Z","signature_b64":"KKAGSGpgeIx8OVMhAV/20/FYMGf/r89y3g2fmz7TYKEbm3xNmgPSe7LuzNlZU1MK5ZLREjCr+w+vinIZyjdkBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"40a7f0e187de6bdbe6e0af62d53ad0e4a69f66cbce1288b229d63e88c01cc0b6","last_reissued_at":"2026-07-05T04:38:50.627281Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:38:50.627281Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Nearly Optimal Algorithms for Linear Contextual Bandits with Adversarial Corruptions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dongruo Zhou, Jiafan He, Quanquan Gu, Tong Zhang","submitted_at":"2022-05-13T17:58:58Z","abstract_excerpt":"We study the linear contextual bandit problem in the presence of adversarial corruption, where the reward at each round is corrupted by an adversary, and the corruption level (i.e., the sum of corruption magnitudes over the horizon) is $C\\geq 0$. The best-known algorithms in this setting are limited in that they either are computationally inefficient or require a strong assumption on the corruption, or their regret is at least $C$ times worse than the regret without corruption. In this paper, to overcome these limitations, we propose a new algorithm based on the principle of optimism in the fa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.06811","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/2205.06811/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":"2205.06811","created_at":"2026-07-05T04:38:50.627343+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.06811v2","created_at":"2026-07-05T04:38:50.627343+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.06811","created_at":"2026-07-05T04:38:50.627343+00:00"},{"alias_kind":"pith_short_12","alias_value":"ICT7BYMH3ZV5","created_at":"2026-07-05T04:38:50.627343+00:00"},{"alias_kind":"pith_short_16","alias_value":"ICT7BYMH3ZV5XZXA","created_at":"2026-07-05T04:38:50.627343+00:00"},{"alias_kind":"pith_short_8","alias_value":"ICT7BYMH","created_at":"2026-07-05T04:38:50.627343+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.08933","citing_title":"Corruption-Tolerant Asynchronous Q-Learning with Near-Optimal Rates","ref_index":39,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S","json":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S.json","graph_json":"https://pith.science/api/pith-number/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/graph.json","events_json":"https://pith.science/api/pith-number/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/events.json","paper":"https://pith.science/paper/ICT7BYMH"},"agent_actions":{"view_html":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S","download_json":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S.json","view_paper":"https://pith.science/paper/ICT7BYMH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.06811&json=true","fetch_graph":"https://pith.science/api/pith-number/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/graph.json","fetch_events":"https://pith.science/api/pith-number/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/action/storage_attestation","attest_author":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/action/author_attestation","sign_citation":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/action/citation_signature","submit_replication":"https://pith.science/pith/ICT7BYMH3ZV5XZXAV5RNKOWQ4S/action/replication_record"}},"created_at":"2026-07-05T04:38:50.627343+00:00","updated_at":"2026-07-05T04:38:50.627343+00:00"}