{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:AGX46EE7ACU5XPWBDCD3GIQ7XY","short_pith_number":"pith:AGX46EE7","schema_version":"1.0","canonical_sha256":"01afcf109f00a9dbbec11887b3221fbe31700f72e221b25ac373b50808a738c4","source":{"kind":"arxiv","id":"2505.21790","version":1},"attestation_state":"computed","paper":{"title":"Faster Rates for Private Adversarial Bandits","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hilal Asi, Kunal Talwar, Vinod Raman","submitted_at":"2025-05-27T21:42:10Z","abstract_excerpt":"We design new differentially private algorithms for the problems of adversarial bandits and bandits with expert advice. For adversarial bandits, we give a simple and efficient conversion of any non-private bandit algorithm to a private bandit algorithm. Instantiating our conversion with existing non-private bandit algorithms gives a regret upper bound of $O\\left(\\frac{\\sqrt{KT}}{\\sqrt{\\epsilon}}\\right)$, improving upon the existing upper bound $O\\left(\\frac{\\sqrt{KT \\log(KT)}}{\\epsilon}\\right)$ for all $\\epsilon \\leq 1$. In particular, our algorithms allow for sublinear expected regret even wh"},"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":"2505.21790","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-27T21:42:10Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c46b359752c0dc5ca0215002db20f543de7b63b24a25c0eeda59d7d319fd8f1c","abstract_canon_sha256":"e847881190d98ee70f4b9606696d23ead47c918df642cf39ff9a7079e03a08f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:03.600290Z","signature_b64":"rbUOxvX2VcfxvzBy6F2y2kavV2vOvgTZmyxuYXlUgpXR2tN7bF3FmvOVNXW7GLUoC8Gs+EdiFf1u96y7R3YXCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"01afcf109f00a9dbbec11887b3221fbe31700f72e221b25ac373b50808a738c4","last_reissued_at":"2026-07-05T11:11:03.599792Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:03.599792Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Faster Rates for Private Adversarial Bandits","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hilal Asi, Kunal Talwar, Vinod Raman","submitted_at":"2025-05-27T21:42:10Z","abstract_excerpt":"We design new differentially private algorithms for the problems of adversarial bandits and bandits with expert advice. For adversarial bandits, we give a simple and efficient conversion of any non-private bandit algorithm to a private bandit algorithm. Instantiating our conversion with existing non-private bandit algorithms gives a regret upper bound of $O\\left(\\frac{\\sqrt{KT}}{\\sqrt{\\epsilon}}\\right)$, improving upon the existing upper bound $O\\left(\\frac{\\sqrt{KT \\log(KT)}}{\\epsilon}\\right)$ for all $\\epsilon \\leq 1$. In particular, our algorithms allow for sublinear expected regret even wh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21790","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/2505.21790/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":"2505.21790","created_at":"2026-07-05T11:11:03.599853+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.21790v1","created_at":"2026-07-05T11:11:03.599853+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21790","created_at":"2026-07-05T11:11:03.599853+00:00"},{"alias_kind":"pith_short_12","alias_value":"AGX46EE7ACU5","created_at":"2026-07-05T11:11:03.599853+00:00"},{"alias_kind":"pith_short_16","alias_value":"AGX46EE7ACU5XPWB","created_at":"2026-07-05T11:11:03.599853+00:00"},{"alias_kind":"pith_short_8","alias_value":"AGX46EE7","created_at":"2026-07-05T11:11:03.599853+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05266","citing_title":"Differential Privacy in the Extensive-Form Bandit Problem","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY","json":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY.json","graph_json":"https://pith.science/api/pith-number/AGX46EE7ACU5XPWBDCD3GIQ7XY/graph.json","events_json":"https://pith.science/api/pith-number/AGX46EE7ACU5XPWBDCD3GIQ7XY/events.json","paper":"https://pith.science/paper/AGX46EE7"},"agent_actions":{"view_html":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY","download_json":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY.json","view_paper":"https://pith.science/paper/AGX46EE7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.21790&json=true","fetch_graph":"https://pith.science/api/pith-number/AGX46EE7ACU5XPWBDCD3GIQ7XY/graph.json","fetch_events":"https://pith.science/api/pith-number/AGX46EE7ACU5XPWBDCD3GIQ7XY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY/action/storage_attestation","attest_author":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY/action/author_attestation","sign_citation":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY/action/citation_signature","submit_replication":"https://pith.science/pith/AGX46EE7ACU5XPWBDCD3GIQ7XY/action/replication_record"}},"created_at":"2026-07-05T11:11:03.599853+00:00","updated_at":"2026-07-05T11:11:03.599853+00:00"}