{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RGZD2C57XZCC3MJDL3MXLIP57D","short_pith_number":"pith:RGZD2C57","schema_version":"1.0","canonical_sha256":"89b23d0bbfbe442db1235ed975a1fdf8d6fc08a4a8f36bca0fcb3243fa4f9dfb","source":{"kind":"arxiv","id":"2405.14657","version":2},"attestation_state":"computed","paper":{"title":"Anchor-Based Heteroscedastic Noise for Preferential Bayesian Optimization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Julien Martinelli, Marshal Arijona Sinaga, Samuel Kaski","submitted_at":"2024-05-23T14:55:18Z","abstract_excerpt":"Preferential Bayesian optimization (PBO) learns latent utilities from pairwise comparisons, but most existing methods assume homoscedastic comparison noise. This is inadequate in human-in-the-loop settings, where a user may compare some designs reliably and others only hesitantly. We propose a heteroscedastic noise model for PBO: before optimization, the user provides a small set of reliable examples, called anchors, and a kernel density estimator (KDE) turns these anchors into an input-dependent map of user uncertainty. We incorporate this map into preferential GP surrogates and derive risk-a"},"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":"2405.14657","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-23T14:55:18Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"93e0c7a7ecd24ef1f5378f0b27fd37ee4452c429836c70fbf9a2fe0642f194ae","abstract_canon_sha256":"cdf9c0d8063624153e21f066843be418d1b45adb2626bc8f4c782f1f96f0cf85"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:46.123065Z","signature_b64":"73HBEVTLMmJ10XM5FR6v+CMbogpueC1+6+8T6gqHZJx+AT1a4eGUZ/iSkQwWGzXy/xgdaXit0HOasgs063p1BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89b23d0bbfbe442db1235ed975a1fdf8d6fc08a4a8f36bca0fcb3243fa4f9dfb","last_reissued_at":"2026-05-20T00:02:46.122489Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:46.122489Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Anchor-Based Heteroscedastic Noise for Preferential Bayesian Optimization","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Julien Martinelli, Marshal Arijona Sinaga, Samuel Kaski","submitted_at":"2024-05-23T14:55:18Z","abstract_excerpt":"Preferential Bayesian optimization (PBO) learns latent utilities from pairwise comparisons, but most existing methods assume homoscedastic comparison noise. This is inadequate in human-in-the-loop settings, where a user may compare some designs reliably and others only hesitantly. We propose a heteroscedastic noise model for PBO: before optimization, the user provides a small set of reliable examples, called anchors, and a kernel density estimator (KDE) turns these anchors into an input-dependent map of user uncertainty. We incorporate this map into preferential GP surrogates and derive risk-a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14657","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/2405.14657/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":"2405.14657","created_at":"2026-05-20T00:02:46.122578+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.14657v2","created_at":"2026-05-20T00:02:46.122578+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14657","created_at":"2026-05-20T00:02:46.122578+00:00"},{"alias_kind":"pith_short_12","alias_value":"RGZD2C57XZCC","created_at":"2026-05-20T00:02:46.122578+00:00"},{"alias_kind":"pith_short_16","alias_value":"RGZD2C57XZCC3MJD","created_at":"2026-05-20T00:02:46.122578+00:00"},{"alias_kind":"pith_short_8","alias_value":"RGZD2C57","created_at":"2026-05-20T00:02:46.122578+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2403.11782","citing_title":"A tutorial on learning from preferences and choices with Gaussian Processes","ref_index":117,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D","json":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D.json","graph_json":"https://pith.science/api/pith-number/RGZD2C57XZCC3MJDL3MXLIP57D/graph.json","events_json":"https://pith.science/api/pith-number/RGZD2C57XZCC3MJDL3MXLIP57D/events.json","paper":"https://pith.science/paper/RGZD2C57"},"agent_actions":{"view_html":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D","download_json":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D.json","view_paper":"https://pith.science/paper/RGZD2C57","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.14657&json=true","fetch_graph":"https://pith.science/api/pith-number/RGZD2C57XZCC3MJDL3MXLIP57D/graph.json","fetch_events":"https://pith.science/api/pith-number/RGZD2C57XZCC3MJDL3MXLIP57D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D/action/storage_attestation","attest_author":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D/action/author_attestation","sign_citation":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D/action/citation_signature","submit_replication":"https://pith.science/pith/RGZD2C57XZCC3MJDL3MXLIP57D/action/replication_record"}},"created_at":"2026-05-20T00:02:46.122578+00:00","updated_at":"2026-05-20T00:02:46.122578+00:00"}