{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:IOQTZG6XY7S7ZMB2R4VIQVXHJK","short_pith_number":"pith:IOQTZG6X","schema_version":"1.0","canonical_sha256":"43a13c9bd7c7e5fcb03a8f2a8856e74abf13cc1dde4fe83ec9d2f3bb16f5b95a","source":{"kind":"arxiv","id":"2206.08873","version":2},"attestation_state":"computed","paper":{"title":"Mirror Descent with Relative Smoothness in Measure Spaces, with application to Sinkhorn and EM","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Anna Korba, Flavien L\\'eger, Pierre-Cyril Aubin-Frankowski","submitted_at":"2022-06-17T16:19:47Z","abstract_excerpt":"Many problems in machine learning can be formulated as optimizing a convex functional over a vector space of measures. This paper studies the convergence of the mirror descent algorithm in this infinite-dimensional setting. Defining Bregman divergences through directional derivatives, we derive the convergence of the scheme for relatively smooth and convex pairs of functionals. Such assumptions allow to handle non-smooth functionals such as the Kullback--Leibler (KL) divergence. Applying our result to joint distributions and KL, we show that Sinkhorn's primal iterations for entropic optimal tr"},"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":"2206.08873","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"math.OC","submitted_at":"2022-06-17T16:19:47Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"f64e49034f9d0db19d6f5b8edc6705463b8f80375af224e50fe199b5f168823a","abstract_canon_sha256":"41d0cc4fca7d09f2ab16985b92885066e45985cb6d2b247b878e147e54403d8d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:05:06.565172Z","signature_b64":"iv3x9pqVoKBhhyiyG1WMMgeBrwMvvhL78wIDnd5F04OO98VJpGN8tupZ7xFxxhUpeAKh5fJciwjDvOXybMHPAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"43a13c9bd7c7e5fcb03a8f2a8856e74abf13cc1dde4fe83ec9d2f3bb16f5b95a","last_reissued_at":"2026-07-05T05:05:06.564695Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:05:06.564695Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mirror Descent with Relative Smoothness in Measure Spaces, with application to Sinkhorn and EM","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Anna Korba, Flavien L\\'eger, Pierre-Cyril Aubin-Frankowski","submitted_at":"2022-06-17T16:19:47Z","abstract_excerpt":"Many problems in machine learning can be formulated as optimizing a convex functional over a vector space of measures. This paper studies the convergence of the mirror descent algorithm in this infinite-dimensional setting. Defining Bregman divergences through directional derivatives, we derive the convergence of the scheme for relatively smooth and convex pairs of functionals. Such assumptions allow to handle non-smooth functionals such as the Kullback--Leibler (KL) divergence. Applying our result to joint distributions and KL, we show that Sinkhorn's primal iterations for entropic optimal tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.08873","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/2206.08873/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":"2206.08873","created_at":"2026-07-05T05:05:06.564754+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.08873v2","created_at":"2026-07-05T05:05:06.564754+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.08873","created_at":"2026-07-05T05:05:06.564754+00:00"},{"alias_kind":"pith_short_12","alias_value":"IOQTZG6XY7S7","created_at":"2026-07-05T05:05:06.564754+00:00"},{"alias_kind":"pith_short_16","alias_value":"IOQTZG6XY7S7ZMB2","created_at":"2026-07-05T05:05:06.564754+00:00"},{"alias_kind":"pith_short_8","alias_value":"IOQTZG6X","created_at":"2026-07-05T05:05:06.564754+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19611","citing_title":"Bregman-projected mirror methods for regularized stationary mean-field games","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK","json":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK.json","graph_json":"https://pith.science/api/pith-number/IOQTZG6XY7S7ZMB2R4VIQVXHJK/graph.json","events_json":"https://pith.science/api/pith-number/IOQTZG6XY7S7ZMB2R4VIQVXHJK/events.json","paper":"https://pith.science/paper/IOQTZG6X"},"agent_actions":{"view_html":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK","download_json":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK.json","view_paper":"https://pith.science/paper/IOQTZG6X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.08873&json=true","fetch_graph":"https://pith.science/api/pith-number/IOQTZG6XY7S7ZMB2R4VIQVXHJK/graph.json","fetch_events":"https://pith.science/api/pith-number/IOQTZG6XY7S7ZMB2R4VIQVXHJK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK/action/storage_attestation","attest_author":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK/action/author_attestation","sign_citation":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK/action/citation_signature","submit_replication":"https://pith.science/pith/IOQTZG6XY7S7ZMB2R4VIQVXHJK/action/replication_record"}},"created_at":"2026-07-05T05:05:06.564754+00:00","updated_at":"2026-07-05T05:05:06.564754+00:00"}