{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:B7CMY7Y4NMUUP7DLFYGNUG4FVE","short_pith_number":"pith:B7CMY7Y4","schema_version":"1.0","canonical_sha256":"0fc4cc7f1c6b2947fc6b2e0cda1b85a92d939252d993e83ac34ab5b9698c01fd","source":{"kind":"arxiv","id":"2207.00736","version":2},"attestation_state":"computed","paper":{"title":"Exponential Convergence of Sinkhorn Under Regularization Scheduling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DS","authors_text":"Arvind Ramaswami, Jingbang Chen, Li Chen, Richard Peng, Yang P. Liu","submitted_at":"2022-07-02T04:30:54Z","abstract_excerpt":"In 2013, Cuturi [Cut13] introduced the Sinkhorn algorithm for matrix scaling as a method to compute solutions to regularized optimal transport problems. In this paper, aiming at a better convergence rate for a high accuracy solution, we work on understanding the Sinkhorn algorithm under regularization scheduling, and thus modify it with a mechanism that adaptively doubles the regularization parameter $\\eta$ periodically. We prove that such modified version of Sinkhorn has an exponential convergence rate as iteration complexity depending on $\\log(1/\\varepsilon)$ instead of $\\varepsilon^{-O(1)}$"},"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":"2207.00736","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DS","submitted_at":"2022-07-02T04:30:54Z","cross_cats_sorted":[],"title_canon_sha256":"aa7ac9c3480d4367d12686f8ef5ca871a52caafdef088f9ea458f5f1bb1c55b6","abstract_canon_sha256":"8a7287fc4f047e22f03d888041f1df42d5aee87f86baf31bc9e5222626908482"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:58:16.312148Z","signature_b64":"FfjT3re310ap32Ou6uCvNosUssyBwU/f1+z6tVQx7FhumqtknaY6/L60nnwnrDAsPYEqlHJj9wQ3NdCNCCnsBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0fc4cc7f1c6b2947fc6b2e0cda1b85a92d939252d993e83ac34ab5b9698c01fd","last_reissued_at":"2026-07-05T05:58:16.311669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:58:16.311669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exponential Convergence of Sinkhorn Under Regularization Scheduling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DS","authors_text":"Arvind Ramaswami, Jingbang Chen, Li Chen, Richard Peng, Yang P. Liu","submitted_at":"2022-07-02T04:30:54Z","abstract_excerpt":"In 2013, Cuturi [Cut13] introduced the Sinkhorn algorithm for matrix scaling as a method to compute solutions to regularized optimal transport problems. In this paper, aiming at a better convergence rate for a high accuracy solution, we work on understanding the Sinkhorn algorithm under regularization scheduling, and thus modify it with a mechanism that adaptively doubles the regularization parameter $\\eta$ periodically. We prove that such modified version of Sinkhorn has an exponential convergence rate as iteration complexity depending on $\\log(1/\\varepsilon)$ instead of $\\varepsilon^{-O(1)}$"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2207.00736","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/2207.00736/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":"2207.00736","created_at":"2026-07-05T05:58:16.311744+00:00"},{"alias_kind":"arxiv_version","alias_value":"2207.00736v2","created_at":"2026-07-05T05:58:16.311744+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2207.00736","created_at":"2026-07-05T05:58:16.311744+00:00"},{"alias_kind":"pith_short_12","alias_value":"B7CMY7Y4NMUU","created_at":"2026-07-05T05:58:16.311744+00:00"},{"alias_kind":"pith_short_16","alias_value":"B7CMY7Y4NMUUP7DL","created_at":"2026-07-05T05:58:16.311744+00:00"},{"alias_kind":"pith_short_8","alias_value":"B7CMY7Y4","created_at":"2026-07-05T05:58:16.311744+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/B7CMY7Y4NMUUP7DLFYGNUG4FVE","json":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE.json","graph_json":"https://pith.science/api/pith-number/B7CMY7Y4NMUUP7DLFYGNUG4FVE/graph.json","events_json":"https://pith.science/api/pith-number/B7CMY7Y4NMUUP7DLFYGNUG4FVE/events.json","paper":"https://pith.science/paper/B7CMY7Y4"},"agent_actions":{"view_html":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE","download_json":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE.json","view_paper":"https://pith.science/paper/B7CMY7Y4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2207.00736&json=true","fetch_graph":"https://pith.science/api/pith-number/B7CMY7Y4NMUUP7DLFYGNUG4FVE/graph.json","fetch_events":"https://pith.science/api/pith-number/B7CMY7Y4NMUUP7DLFYGNUG4FVE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE/action/storage_attestation","attest_author":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE/action/author_attestation","sign_citation":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE/action/citation_signature","submit_replication":"https://pith.science/pith/B7CMY7Y4NMUUP7DLFYGNUG4FVE/action/replication_record"}},"created_at":"2026-07-05T05:58:16.311744+00:00","updated_at":"2026-07-05T05:58:16.311744+00:00"}