{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KXII6DFQJMR74WY4DBDA6ZUV27","short_pith_number":"pith:KXII6DFQ","schema_version":"1.0","canonical_sha256":"55d08f0cb04b23fe5b1c18460f6695d7eb15f5106b7aabfc532d1f9f00124da9","source":{"kind":"arxiv","id":"2603.09276","version":2},"attestation_state":"computed","paper":{"title":"On Regret Bounds of Thompson Sampling for Bayesian Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Shion Takeno, Shogo Iwazaki","submitted_at":"2026-03-10T07:02:20Z","abstract_excerpt":"We study a widely used Bayesian optimization method, Gaussian process Thompson sampling (GP-TS), under the assumption that the objective function is a sample path from a GP. Compared with the GP upper confidence bound (GP-UCB) with established high-probability and expected regret bounds, most analyses of GP-TS have been limited to expected regret. Moreover, whether the recent analyses of GP-UCB for the lenient regret and the improved cumulative regret upper bound can be applied to GP-TS remains unclear. To fill these gaps, this paper shows several regret bounds: (i) a regret lower bound for GP"},"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":"2603.09276","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2026-03-10T07:02:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4393967a6b88f56d20674edb8eea6fa76960c5420b610a03906aff5815b22977","abstract_canon_sha256":"a040cf5af1565b3cdffc14a67a04f0f7f40ea9247c7ce8faf40f5d53c7d09759"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:09:32.621626Z","signature_b64":"HAu8kP7S8+ZuAB9AMXI8nvHxnGIkGeclQbtCSrBt/GmX3shCk5jUaGrKA/w5XFiqfU258XIvbrXKY6J2sDkvCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55d08f0cb04b23fe5b1c18460f6695d7eb15f5106b7aabfc532d1f9f00124da9","last_reissued_at":"2026-06-11T01:09:32.620552Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:09:32.620552Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Regret Bounds of Thompson Sampling for Bayesian Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Shion Takeno, Shogo Iwazaki","submitted_at":"2026-03-10T07:02:20Z","abstract_excerpt":"We study a widely used Bayesian optimization method, Gaussian process Thompson sampling (GP-TS), under the assumption that the objective function is a sample path from a GP. Compared with the GP upper confidence bound (GP-UCB) with established high-probability and expected regret bounds, most analyses of GP-TS have been limited to expected regret. Moreover, whether the recent analyses of GP-UCB for the lenient regret and the improved cumulative regret upper bound can be applied to GP-TS remains unclear. To fill these gaps, this paper shows several regret bounds: (i) a regret lower bound for GP"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.09276","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/2603.09276/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":"2603.09276","created_at":"2026-06-11T01:09:32.620702+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.09276v2","created_at":"2026-06-11T01:09:32.620702+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.09276","created_at":"2026-06-11T01:09:32.620702+00:00"},{"alias_kind":"pith_short_12","alias_value":"KXII6DFQJMR7","created_at":"2026-06-11T01:09:32.620702+00:00"},{"alias_kind":"pith_short_16","alias_value":"KXII6DFQJMR74WY4","created_at":"2026-06-11T01:09:32.620702+00:00"},{"alias_kind":"pith_short_8","alias_value":"KXII6DFQ","created_at":"2026-06-11T01:09:32.620702+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.10299","citing_title":"Nearly-Optimal Algorithm for Adversarial Kernelized Bandits","ref_index":62,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10299","citing_title":"Nearly-Optimal Algorithm for Adversarial Kernelized Bandits","ref_index":62,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27","json":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27.json","graph_json":"https://pith.science/api/pith-number/KXII6DFQJMR74WY4DBDA6ZUV27/graph.json","events_json":"https://pith.science/api/pith-number/KXII6DFQJMR74WY4DBDA6ZUV27/events.json","paper":"https://pith.science/paper/KXII6DFQ"},"agent_actions":{"view_html":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27","download_json":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27.json","view_paper":"https://pith.science/paper/KXII6DFQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.09276&json=true","fetch_graph":"https://pith.science/api/pith-number/KXII6DFQJMR74WY4DBDA6ZUV27/graph.json","fetch_events":"https://pith.science/api/pith-number/KXII6DFQJMR74WY4DBDA6ZUV27/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27/action/storage_attestation","attest_author":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27/action/author_attestation","sign_citation":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27/action/citation_signature","submit_replication":"https://pith.science/pith/KXII6DFQJMR74WY4DBDA6ZUV27/action/replication_record"}},"created_at":"2026-06-11T01:09:32.620702+00:00","updated_at":"2026-06-11T01:09:32.620702+00:00"}