{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:RVQDQLZTNF5O4FOSXSYAZWLLBG","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"cc0de02cd21186acf6e9e5d493a1a5768873148fe55d7a407448af19180c8ebf","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-11T12:53:38Z","title_canon_sha256":"ef86c2fdb573021cc2987481c740bc25c99711b3e57de7cbd3f9754e8f78ab32"},"schema_version":"1.0","source":{"id":"1906.04540","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.04540","created_at":"2026-07-05T01:51:04Z"},{"alias_kind":"arxiv_version","alias_value":"1906.04540v3","created_at":"2026-07-05T01:51:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.04540","created_at":"2026-07-05T01:51:04Z"},{"alias_kind":"pith_short_12","alias_value":"RVQDQLZTNF5O","created_at":"2026-07-05T01:51:04Z"},{"alias_kind":"pith_short_16","alias_value":"RVQDQLZTNF5O4FOS","created_at":"2026-07-05T01:51:04Z"},{"alias_kind":"pith_short_8","alias_value":"RVQDQLZT","created_at":"2026-07-05T01:51:04Z"}],"graph_snapshots":[{"event_id":"sha256:72fa30902733d2aedc0393eef78e6fd00458a047a30e6c120e494c0c006151f5","target":"graph","created_at":"2026-07-05T01:51:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1906.04540/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper shows that the implicit bias of gradient descent on linearly separable data is exactly characterized by the optimal solution of a dual optimization problem given by a smoothed margin, even for general losses. This is in contrast to prior results, which are often tailored to exponentially-tailed losses. For the exponential loss specifically, with $n$ training examples and $t$ gradient descent steps, our dual analysis further allows us to prove an $O(\\ln(n)/\\ln(t))$ convergence rate to the $\\ell_2$ maximum margin direction, when a constant step size is used. This rate is tight in both","authors_text":"Matus Telgarsky, Ziwei Ji","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-11T12:53:38Z","title":"Characterizing the implicit bias via a primal-dual analysis"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.04540","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:823328cb6be2f787e1b16dfcc96b14cd89c86e24d8d0ec8b7efaac6371688a83","target":"record","created_at":"2026-07-05T01:51:04Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"cc0de02cd21186acf6e9e5d493a1a5768873148fe55d7a407448af19180c8ebf","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-11T12:53:38Z","title_canon_sha256":"ef86c2fdb573021cc2987481c740bc25c99711b3e57de7cbd3f9754e8f78ab32"},"schema_version":"1.0","source":{"id":"1906.04540","kind":"arxiv","version":3}},"canonical_sha256":"8d60382f33697aee15d2bcb00cd96b09b1f828c75b82af46eb74c2d3b1275a97","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8d60382f33697aee15d2bcb00cd96b09b1f828c75b82af46eb74c2d3b1275a97","first_computed_at":"2026-07-05T01:51:04.764988Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:51:04.764988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"enq53oZZxQGCpFgaepIPQKB5djj4e/KYeKz7OJMeUrTBnJlSKeEmVCfTNIEh238PYoXJblYGTZZoVRmPDt4dBA==","signature_status":"signed_v1","signed_at":"2026-07-05T01:51:04.765391Z","signed_message":"canonical_sha256_bytes"},"source_id":"1906.04540","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:823328cb6be2f787e1b16dfcc96b14cd89c86e24d8d0ec8b7efaac6371688a83","sha256:72fa30902733d2aedc0393eef78e6fd00458a047a30e6c120e494c0c006151f5"],"state_sha256":"36fda6251ef61a27687626466793e17899547331f2283cbe171ad032df8d8d86"}