{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:SN6EPQSZUK6VQ5ADNNBJ3C5TFR","short_pith_number":"pith:SN6EPQSZ","canonical_record":{"source":{"id":"1904.12443","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2019-04-29T04:33:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2d082f11c9f27fb98bba25fb955788c8439e8eb17f5f420837df170fd0c9c594","abstract_canon_sha256":"c4330f8aa2e2c259cb01632565eec2b26e8a43ac6552f54075cf9aa6438cace6"},"schema_version":"1.0"},"canonical_sha256":"937c47c259a2bd5874036b429d8bb32c488fd0475e0c85da17a8be10436b6cbe","source":{"kind":"arxiv","id":"1904.12443","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1904.12443","created_at":"2026-05-17T23:44:47Z"},{"alias_kind":"arxiv_version","alias_value":"1904.12443v2","created_at":"2026-05-17T23:44:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.12443","created_at":"2026-05-17T23:44:47Z"},{"alias_kind":"pith_short_12","alias_value":"SN6EPQSZUK6V","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_16","alias_value":"SN6EPQSZUK6VQ5AD","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_8","alias_value":"SN6EPQSZ","created_at":"2026-05-18T12:33:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:SN6EPQSZUK6VQ5ADNNBJ3C5TFR","target":"record","payload":{"canonical_record":{"source":{"id":"1904.12443","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2019-04-29T04:33:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2d082f11c9f27fb98bba25fb955788c8439e8eb17f5f420837df170fd0c9c594","abstract_canon_sha256":"c4330f8aa2e2c259cb01632565eec2b26e8a43ac6552f54075cf9aa6438cace6"},"schema_version":"1.0"},"canonical_sha256":"937c47c259a2bd5874036b429d8bb32c488fd0475e0c85da17a8be10436b6cbe","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:44:47.265127Z","signature_b64":"eSTut8SlyhWKhaf7A2Pl/3Atcst30e78QLmgwb98oML4gVTWQxJPnKHTEw0c8fO/VcqYIOp/BZDCkdebxOwtBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"937c47c259a2bd5874036b429d8bb32c488fd0475e0c85da17a8be10436b6cbe","last_reissued_at":"2026-05-17T23:44:47.264701Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:44:47.264701Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1904.12443","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:44:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"c2aTHlvTksXej+M5oYrkbM3FQwaLuKuLv+2jDVsBfpLLCHXHvYtUulco6KWOs/AnqXT1tbM3uEdkHcpeqmGfBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T07:17:28.887673Z"},"content_sha256":"039ce592301f5cf134317b439e51537ced21b1fcde916a632cbd937c938d9fde","schema_version":"1.0","event_id":"sha256:039ce592301f5cf134317b439e51537ced21b1fcde916a632cbd937c938d9fde"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:SN6EPQSZUK6VQ5ADNNBJ3C5TFR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Making the Last Iterate of SGD Information Theoretically Optimal","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"math.OC","authors_text":"Dheeraj Nagaraj, Praneeth Netrapalli, Prateek Jain","submitted_at":"2019-04-29T04:33:53Z","abstract_excerpt":"Stochastic gradient descent (SGD) is one of the most widely used algorithms for large scale optimization problems. While classical theoretical analysis of SGD for convex problems studies (suffix) \\emph{averages} of iterates and obtains information theoretically optimal bounds on suboptimality, the \\emph{last point} of SGD is, by far, the most preferred choice in practice. The best known results for last point of SGD \\cite{shamir2013stochastic} however, are suboptimal compared to information theoretic lower bounds by a $\\log T$ factor, where $T$ is the number of iterations. \\cite{harvey2018tigh"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.12443","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":""},"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":"2026-05-17T23:44:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"glTpUTzcn4BoDou3h15V9k8LZHWu4kgTypj/83FGCXuAdi9Xw/GLpM7B+2bnm6AgcLfuZ4kjyLgguFIAp83PDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T07:17:28.888017Z"},"content_sha256":"ae4e25b92ce13944601aa48a7bb1756c407874a3f45dfc41921866e401ff474d","schema_version":"1.0","event_id":"sha256:ae4e25b92ce13944601aa48a7bb1756c407874a3f45dfc41921866e401ff474d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SN6EPQSZUK6VQ5ADNNBJ3C5TFR/bundle.json","state_url":"https://pith.science/pith/SN6EPQSZUK6VQ5ADNNBJ3C5TFR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SN6EPQSZUK6VQ5ADNNBJ3C5TFR/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":"2026-05-28T07:17:28Z","links":{"resolver":"https://pith.science/pith/SN6EPQSZUK6VQ5ADNNBJ3C5TFR","bundle":"https://pith.science/pith/SN6EPQSZUK6VQ5ADNNBJ3C5TFR/bundle.json","state":"https://pith.science/pith/SN6EPQSZUK6VQ5ADNNBJ3C5TFR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SN6EPQSZUK6VQ5ADNNBJ3C5TFR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:SN6EPQSZUK6VQ5ADNNBJ3C5TFR","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":"c4330f8aa2e2c259cb01632565eec2b26e8a43ac6552f54075cf9aa6438cace6","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2019-04-29T04:33:53Z","title_canon_sha256":"2d082f11c9f27fb98bba25fb955788c8439e8eb17f5f420837df170fd0c9c594"},"schema_version":"1.0","source":{"id":"1904.12443","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1904.12443","created_at":"2026-05-17T23:44:47Z"},{"alias_kind":"arxiv_version","alias_value":"1904.12443v2","created_at":"2026-05-17T23:44:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.12443","created_at":"2026-05-17T23:44:47Z"},{"alias_kind":"pith_short_12","alias_value":"SN6EPQSZUK6V","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_16","alias_value":"SN6EPQSZUK6VQ5AD","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_8","alias_value":"SN6EPQSZ","created_at":"2026-05-18T12:33:27Z"}],"graph_snapshots":[{"event_id":"sha256:ae4e25b92ce13944601aa48a7bb1756c407874a3f45dfc41921866e401ff474d","target":"graph","created_at":"2026-05-17T23:44:47Z","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":"Stochastic gradient descent (SGD) is one of the most widely used algorithms for large scale optimization problems. While classical theoretical analysis of SGD for convex problems studies (suffix) \\emph{averages} of iterates and obtains information theoretically optimal bounds on suboptimality, the \\emph{last point} of SGD is, by far, the most preferred choice in practice. The best known results for last point of SGD \\cite{shamir2013stochastic} however, are suboptimal compared to information theoretic lower bounds by a $\\log T$ factor, where $T$ is the number of iterations. \\cite{harvey2018tigh","authors_text":"Dheeraj Nagaraj, Praneeth Netrapalli, Prateek Jain","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2019-04-29T04:33:53Z","title":"Making the Last Iterate of SGD Information Theoretically Optimal"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.12443","kind":"arxiv","version":2},"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":"sha256:039ce592301f5cf134317b439e51537ced21b1fcde916a632cbd937c938d9fde","target":"record","created_at":"2026-05-17T23:44:47Z","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":"c4330f8aa2e2c259cb01632565eec2b26e8a43ac6552f54075cf9aa6438cace6","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2019-04-29T04:33:53Z","title_canon_sha256":"2d082f11c9f27fb98bba25fb955788c8439e8eb17f5f420837df170fd0c9c594"},"schema_version":"1.0","source":{"id":"1904.12443","kind":"arxiv","version":2}},"canonical_sha256":"937c47c259a2bd5874036b429d8bb32c488fd0475e0c85da17a8be10436b6cbe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"937c47c259a2bd5874036b429d8bb32c488fd0475e0c85da17a8be10436b6cbe","first_computed_at":"2026-05-17T23:44:47.264701Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:44:47.264701Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eSTut8SlyhWKhaf7A2Pl/3Atcst30e78QLmgwb98oML4gVTWQxJPnKHTEw0c8fO/VcqYIOp/BZDCkdebxOwtBA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:44:47.265127Z","signed_message":"canonical_sha256_bytes"},"source_id":"1904.12443","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:039ce592301f5cf134317b439e51537ced21b1fcde916a632cbd937c938d9fde","sha256:ae4e25b92ce13944601aa48a7bb1756c407874a3f45dfc41921866e401ff474d"],"state_sha256":"e43157a57ffd8abf21e3bf448f0e6c67b95b86fe8f10ab5d832d70182c699a9e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IFtKo/Q5580O+LPJszI10ziDk9+TbobQR9PrrprwEqT+COJujaWDYv8SuaqX3G3FZGtvv5r80VvGkn8NX48rCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T07:17:28.889975Z","bundle_sha256":"5a26d8beea5cc50cfe13a3a68af59ef5b8c16c205928176afe0cfb223390c405"}}