{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:PYRHPLAAOK2FLHW7EE4S3Y7F2C","short_pith_number":"pith:PYRHPLAA","schema_version":"1.0","canonical_sha256":"7e2277ac0072b4559edf21392de3e5d0a01419d057b1d3491f99dcd9bced9492","source":{"kind":"arxiv","id":"2302.01332","version":2},"attestation_state":"computed","paper":{"title":"Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Frederik Warburg, Marco Miani, Silas Brack, Soren Hauberg","submitted_at":"2023-02-02T18:59:23Z","abstract_excerpt":"We propose the first Bayesian encoder for metric learning. Rather than relying on neural amortization as done in prior works, we learn a distribution over the network weights with the Laplace Approximation. We actualize this by first proving that the contrastive loss is a valid log-posterior. We then propose three methods that ensure a positive definite Hessian. Lastly, we present a novel decomposition of the Generalized Gauss-Newton approximation. Empirically, we show that our Laplacian Metric Learner (LAM) estimates well-calibrated uncertainties, reliably detects out-of-distribution examples"},"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":"2302.01332","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-02-02T18:59:23Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"6ec57669bbb7ce550be1ee7035b759ade4d3925138c024fd365999db59462536","abstract_canon_sha256":"34dafa02af07bbb827dd580f154bfff1fac590feddcdc7ac8b4c9b8cdbaf414f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:38:47.069535Z","signature_b64":"kjVToCGQXnVb19zgGu/yascxlQXGVY8B7icbdssQj4+lnUUc6iTLShznPhWryunAdr5yKe7fwAMFSzLjA1llBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e2277ac0072b4559edf21392de3e5d0a01419d057b1d3491f99dcd9bced9492","last_reissued_at":"2026-07-05T05:38:47.068919Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:38:47.068919Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bayesian Metric Learning for Uncertainty Quantification in Image Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Frederik Warburg, Marco Miani, Silas Brack, Soren Hauberg","submitted_at":"2023-02-02T18:59:23Z","abstract_excerpt":"We propose the first Bayesian encoder for metric learning. Rather than relying on neural amortization as done in prior works, we learn a distribution over the network weights with the Laplace Approximation. We actualize this by first proving that the contrastive loss is a valid log-posterior. We then propose three methods that ensure a positive definite Hessian. Lastly, we present a novel decomposition of the Generalized Gauss-Newton approximation. Empirically, we show that our Laplacian Metric Learner (LAM) estimates well-calibrated uncertainties, reliably detects out-of-distribution examples"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.01332","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/2302.01332/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":"2302.01332","created_at":"2026-07-05T05:38:47.069005+00:00"},{"alias_kind":"arxiv_version","alias_value":"2302.01332v2","created_at":"2026-07-05T05:38:47.069005+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.01332","created_at":"2026-07-05T05:38:47.069005+00:00"},{"alias_kind":"pith_short_12","alias_value":"PYRHPLAAOK2F","created_at":"2026-07-05T05:38:47.069005+00:00"},{"alias_kind":"pith_short_16","alias_value":"PYRHPLAAOK2FLHW7","created_at":"2026-07-05T05:38:47.069005+00:00"},{"alias_kind":"pith_short_8","alias_value":"PYRHPLAA","created_at":"2026-07-05T05:38:47.069005+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C","json":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C.json","graph_json":"https://pith.science/api/pith-number/PYRHPLAAOK2FLHW7EE4S3Y7F2C/graph.json","events_json":"https://pith.science/api/pith-number/PYRHPLAAOK2FLHW7EE4S3Y7F2C/events.json","paper":"https://pith.science/paper/PYRHPLAA"},"agent_actions":{"view_html":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C","download_json":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C.json","view_paper":"https://pith.science/paper/PYRHPLAA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2302.01332&json=true","fetch_graph":"https://pith.science/api/pith-number/PYRHPLAAOK2FLHW7EE4S3Y7F2C/graph.json","fetch_events":"https://pith.science/api/pith-number/PYRHPLAAOK2FLHW7EE4S3Y7F2C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C/action/storage_attestation","attest_author":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C/action/author_attestation","sign_citation":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C/action/citation_signature","submit_replication":"https://pith.science/pith/PYRHPLAAOK2FLHW7EE4S3Y7F2C/action/replication_record"}},"created_at":"2026-07-05T05:38:47.069005+00:00","updated_at":"2026-07-05T05:38:47.069005+00:00"}