{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:BBQUBRXQREJUVAJVNOKZAYBACP","short_pith_number":"pith:BBQUBRXQ","canonical_record":{"source":{"id":"1811.02564","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-11-06T00:05:00Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"be19f9e842d0ea94c2409c70eaab45a86d023465f73684ad95dc1b1feff6f6cd","abstract_canon_sha256":"469239f4283212000190cc6e1b26d2f0f2e6f6e64a8bfef9a2946421d3222dbc"},"schema_version":"1.0"},"canonical_sha256":"086140c6f089134a81356b9590602013fa436e6b2f906a63408a99b113eb13a2","source":{"kind":"arxiv","id":"1811.02564","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.02564","created_at":"2026-05-18T00:01:21Z"},{"alias_kind":"arxiv_version","alias_value":"1811.02564v1","created_at":"2026-05-18T00:01:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.02564","created_at":"2026-05-18T00:01:21Z"},{"alias_kind":"pith_short_12","alias_value":"BBQUBRXQREJU","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"BBQUBRXQREJUVAJV","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"BBQUBRXQ","created_at":"2026-05-18T12:32:13Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:BBQUBRXQREJUVAJVNOKZAYBACP","target":"record","payload":{"canonical_record":{"source":{"id":"1811.02564","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-11-06T00:05:00Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"be19f9e842d0ea94c2409c70eaab45a86d023465f73684ad95dc1b1feff6f6cd","abstract_canon_sha256":"469239f4283212000190cc6e1b26d2f0f2e6f6e64a8bfef9a2946421d3222dbc"},"schema_version":"1.0"},"canonical_sha256":"086140c6f089134a81356b9590602013fa436e6b2f906a63408a99b113eb13a2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:01:21.332051Z","signature_b64":"eJLP6E06YuQ/nxOcl6noI8/psqas3DxEeb0B1N+B2jvYU2nlB2zWQLJhE5r6gq1+66dsnI2IUsKZTYOYXM7NDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"086140c6f089134a81356b9590602013fa436e6b2f906a63408a99b113eb13a2","last_reissued_at":"2026-05-18T00:01:21.331516Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:01:21.331516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1811.02564","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:01:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5GWdOxCbHQrk1NHm/zd3vvb1+ZrYguMEUCjDdt1btdLHxpCuILkx7vMmswJ9A8xYZd4SUvv4hrInnVbrA05NAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T01:50:53.729367Z"},"content_sha256":"bdcb71b04cdcc66454b721a22e76ccf617a314b2e5a200ad71a688088ef2384d","schema_version":"1.0","event_id":"sha256:bdcb71b04cdcc66454b721a22e76ccf617a314b2e5a200ad71a688088ef2384d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:BBQUBRXQREJUVAJVNOKZAYBACP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On exponential convergence of SGD in non-convex over-parametrized learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Mikhail Belkin, Raef Bassily, Siyuan Ma","submitted_at":"2018-11-06T00:05:00Z","abstract_excerpt":"Large over-parametrized models learned via stochastic gradient descent (SGD) methods have become a key element in modern machine learning. Although SGD methods are very effective in practice, most theoretical analyses of SGD suggest slower convergence than what is empirically observed. In our recent work [8] we analyzed how interpolation, common in modern over-parametrized learning, results in exponential convergence of SGD with constant step size for convex loss functions. In this note, we extend those results to a much broader non-convex function class satisfying the Polyak-Lojasiewicz (PL) "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.02564","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:01:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CT6xNtOC6RiJifrx+KkLU27zih459yP+PJne1LqhMV3WZh41iq82I1ggGh9iQmKA1GWE70CghGBQeTF63iJ5CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T01:50:53.730036Z"},"content_sha256":"bc78bcb318c88e49a69bd265b746dc3febd7f20755c6dd10ef7ae89e8739351a","schema_version":"1.0","event_id":"sha256:bc78bcb318c88e49a69bd265b746dc3febd7f20755c6dd10ef7ae89e8739351a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BBQUBRXQREJUVAJVNOKZAYBACP/bundle.json","state_url":"https://pith.science/pith/BBQUBRXQREJUVAJVNOKZAYBACP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BBQUBRXQREJUVAJVNOKZAYBACP/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-26T01:50:53Z","links":{"resolver":"https://pith.science/pith/BBQUBRXQREJUVAJVNOKZAYBACP","bundle":"https://pith.science/pith/BBQUBRXQREJUVAJVNOKZAYBACP/bundle.json","state":"https://pith.science/pith/BBQUBRXQREJUVAJVNOKZAYBACP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BBQUBRXQREJUVAJVNOKZAYBACP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:BBQUBRXQREJUVAJVNOKZAYBACP","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":"469239f4283212000190cc6e1b26d2f0f2e6f6e64a8bfef9a2946421d3222dbc","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-11-06T00:05:00Z","title_canon_sha256":"be19f9e842d0ea94c2409c70eaab45a86d023465f73684ad95dc1b1feff6f6cd"},"schema_version":"1.0","source":{"id":"1811.02564","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.02564","created_at":"2026-05-18T00:01:21Z"},{"alias_kind":"arxiv_version","alias_value":"1811.02564v1","created_at":"2026-05-18T00:01:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.02564","created_at":"2026-05-18T00:01:21Z"},{"alias_kind":"pith_short_12","alias_value":"BBQUBRXQREJU","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_16","alias_value":"BBQUBRXQREJUVAJV","created_at":"2026-05-18T12:32:13Z"},{"alias_kind":"pith_short_8","alias_value":"BBQUBRXQ","created_at":"2026-05-18T12:32:13Z"}],"graph_snapshots":[{"event_id":"sha256:bc78bcb318c88e49a69bd265b746dc3febd7f20755c6dd10ef7ae89e8739351a","target":"graph","created_at":"2026-05-18T00:01:21Z","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"},"paper":{"abstract_excerpt":"Large over-parametrized models learned via stochastic gradient descent (SGD) methods have become a key element in modern machine learning. Although SGD methods are very effective in practice, most theoretical analyses of SGD suggest slower convergence than what is empirically observed. In our recent work [8] we analyzed how interpolation, common in modern over-parametrized learning, results in exponential convergence of SGD with constant step size for convex loss functions. In this note, we extend those results to a much broader non-convex function class satisfying the Polyak-Lojasiewicz (PL) ","authors_text":"Mikhail Belkin, Raef Bassily, Siyuan Ma","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-11-06T00:05:00Z","title":"On exponential convergence of SGD in non-convex over-parametrized learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.02564","kind":"arxiv","version":1},"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:bdcb71b04cdcc66454b721a22e76ccf617a314b2e5a200ad71a688088ef2384d","target":"record","created_at":"2026-05-18T00:01:21Z","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":"469239f4283212000190cc6e1b26d2f0f2e6f6e64a8bfef9a2946421d3222dbc","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-11-06T00:05:00Z","title_canon_sha256":"be19f9e842d0ea94c2409c70eaab45a86d023465f73684ad95dc1b1feff6f6cd"},"schema_version":"1.0","source":{"id":"1811.02564","kind":"arxiv","version":1}},"canonical_sha256":"086140c6f089134a81356b9590602013fa436e6b2f906a63408a99b113eb13a2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"086140c6f089134a81356b9590602013fa436e6b2f906a63408a99b113eb13a2","first_computed_at":"2026-05-18T00:01:21.331516Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:01:21.331516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eJLP6E06YuQ/nxOcl6noI8/psqas3DxEeb0B1N+B2jvYU2nlB2zWQLJhE5r6gq1+66dsnI2IUsKZTYOYXM7NDw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:01:21.332051Z","signed_message":"canonical_sha256_bytes"},"source_id":"1811.02564","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:bdcb71b04cdcc66454b721a22e76ccf617a314b2e5a200ad71a688088ef2384d","sha256:bc78bcb318c88e49a69bd265b746dc3febd7f20755c6dd10ef7ae89e8739351a"],"state_sha256":"61cee75e215061818a0762f1437eb7743f00611962a4ed2a7fb90731dbefa7d9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dt+lNEwrHu41XBzPywbTr/m3IvlTGp8hzpzKPW7MCHpjMLpeTyeTA2P4HWw9CXvlS1r8nSu2uIQCt6Nt+vooAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T01:50:53.733495Z","bundle_sha256":"7d447e32f9d8d7b6715bacd1507368edd96c70638c861d50b2c965ad09fe5b96"}}