{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:7ESMEIJMEENZ4WD7HNEDDTM7UE","short_pith_number":"pith:7ESMEIJM","canonical_record":{"source":{"id":"1906.03532","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T22:02:28Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"5281d8d42ff155f0568c4150d75c210db8ee8d9a9b85939f1fbaee841c69cf34","abstract_canon_sha256":"6bcebb28ceda479d15bb7fe80ba761d5f23aa12467bd7558b43533a43fb2db01"},"schema_version":"1.0"},"canonical_sha256":"f924c2212c211b9e587f3b4831cd9fa10605efe3f8351d535b5220a6b2f080d7","source":{"kind":"arxiv","id":"1906.03532","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.03532","created_at":"2026-05-17T23:42:59Z"},{"alias_kind":"arxiv_version","alias_value":"1906.03532v2","created_at":"2026-05-17T23:42:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.03532","created_at":"2026-05-17T23:42:59Z"},{"alias_kind":"pith_short_12","alias_value":"7ESMEIJMEENZ","created_at":"2026-05-18T12:33:12Z"},{"alias_kind":"pith_short_16","alias_value":"7ESMEIJMEENZ4WD7","created_at":"2026-05-18T12:33:12Z"},{"alias_kind":"pith_short_8","alias_value":"7ESMEIJM","created_at":"2026-05-18T12:33:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:7ESMEIJMEENZ4WD7HNEDDTM7UE","target":"record","payload":{"canonical_record":{"source":{"id":"1906.03532","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T22:02:28Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"5281d8d42ff155f0568c4150d75c210db8ee8d9a9b85939f1fbaee841c69cf34","abstract_canon_sha256":"6bcebb28ceda479d15bb7fe80ba761d5f23aa12467bd7558b43533a43fb2db01"},"schema_version":"1.0"},"canonical_sha256":"f924c2212c211b9e587f3b4831cd9fa10605efe3f8351d535b5220a6b2f080d7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:42:59.867690Z","signature_b64":"7vMVZa+9VtMJVKb29YQ+W3Vqm40C+1Pkcac4KM25hDR9g4bSbVOi9chCtgLfi5JWHziK3WOQi9gHW2ri350EDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f924c2212c211b9e587f3b4831cd9fa10605efe3f8351d535b5220a6b2f080d7","last_reissued_at":"2026-05-17T23:42:59.867262Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:42:59.867262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1906.03532","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":"2026-05-17T23:42:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WrxQz83H5kCEBdAPD2ud3qLt6azQeOd9l22oejvAv4X8o2Yqz6MS5YWhB2tdarRXGBETi7gVAT9/dVy2+M+CBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T06:54:53.664403Z"},"content_sha256":"82e2c65a5475635dc02592d414ceefad69c99e2e07d2e176748513f79788a5a8","schema_version":"1.0","event_id":"sha256:82e2c65a5475635dc02592d414ceefad69c99e2e07d2e176748513f79788a5a8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:7ESMEIJMEENZ4WD7HNEDDTM7UE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reducing the variance in online optimization by transporting past gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Ioannis Mitliagkas, Nicolas Le Roux, Pierre-Antoine Manzagol, Reza Babanezhad, S\\'ebastien M. R. Arnold","submitted_at":"2019-06-08T22:02:28Z","abstract_excerpt":"Most stochastic optimization methods use gradients once before discarding them. While variance reduction methods have shown that reusing past gradients can be beneficial when there is a finite number of datapoints, they do not easily extend to the online setting. One issue is the staleness due to using past gradients. We propose to correct this staleness using the idea of implicit gradient transport (IGT) which transforms gradients computed at previous iterates into gradients evaluated at the current iterate without using the Hessian explicitly. In addition to reducing the variance and bias of"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.03532","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":"2026-05-17T23:42:59Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JqO65nDiwwGi9ohn7n2IQWPzKXub0IfFl0XOu5/q8es6Rm+eWLvZjBbW9NqMIVBPcjGxLgVrwaw/Xc4R6XdxDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T06:54:53.664760Z"},"content_sha256":"4f7c75f09448dcc7e9f0d417ebbd3b254ffd2ce5bca0ca4309e8fa7bf06501f3","schema_version":"1.0","event_id":"sha256:4f7c75f09448dcc7e9f0d417ebbd3b254ffd2ce5bca0ca4309e8fa7bf06501f3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7ESMEIJMEENZ4WD7HNEDDTM7UE/bundle.json","state_url":"https://pith.science/pith/7ESMEIJMEENZ4WD7HNEDDTM7UE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7ESMEIJMEENZ4WD7HNEDDTM7UE/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":"2026-05-30T06:54:53Z","links":{"resolver":"https://pith.science/pith/7ESMEIJMEENZ4WD7HNEDDTM7UE","bundle":"https://pith.science/pith/7ESMEIJMEENZ4WD7HNEDDTM7UE/bundle.json","state":"https://pith.science/pith/7ESMEIJMEENZ4WD7HNEDDTM7UE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7ESMEIJMEENZ4WD7HNEDDTM7UE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:7ESMEIJMEENZ4WD7HNEDDTM7UE","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":"6bcebb28ceda479d15bb7fe80ba761d5f23aa12467bd7558b43533a43fb2db01","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T22:02:28Z","title_canon_sha256":"5281d8d42ff155f0568c4150d75c210db8ee8d9a9b85939f1fbaee841c69cf34"},"schema_version":"1.0","source":{"id":"1906.03532","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1906.03532","created_at":"2026-05-17T23:42:59Z"},{"alias_kind":"arxiv_version","alias_value":"1906.03532v2","created_at":"2026-05-17T23:42:59Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.03532","created_at":"2026-05-17T23:42:59Z"},{"alias_kind":"pith_short_12","alias_value":"7ESMEIJMEENZ","created_at":"2026-05-18T12:33:12Z"},{"alias_kind":"pith_short_16","alias_value":"7ESMEIJMEENZ4WD7","created_at":"2026-05-18T12:33:12Z"},{"alias_kind":"pith_short_8","alias_value":"7ESMEIJM","created_at":"2026-05-18T12:33:12Z"}],"graph_snapshots":[{"event_id":"sha256:4f7c75f09448dcc7e9f0d417ebbd3b254ffd2ce5bca0ca4309e8fa7bf06501f3","target":"graph","created_at":"2026-05-17T23:42:59Z","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":"Most stochastic optimization methods use gradients once before discarding them. While variance reduction methods have shown that reusing past gradients can be beneficial when there is a finite number of datapoints, they do not easily extend to the online setting. One issue is the staleness due to using past gradients. We propose to correct this staleness using the idea of implicit gradient transport (IGT) which transforms gradients computed at previous iterates into gradients evaluated at the current iterate without using the Hessian explicitly. In addition to reducing the variance and bias of","authors_text":"Ioannis Mitliagkas, Nicolas Le Roux, Pierre-Antoine Manzagol, Reza Babanezhad, S\\'ebastien M. R. Arnold","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T22:02:28Z","title":"Reducing the variance in online optimization by transporting past gradients"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.03532","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":"sha256:82e2c65a5475635dc02592d414ceefad69c99e2e07d2e176748513f79788a5a8","target":"record","created_at":"2026-05-17T23:42:59Z","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":"6bcebb28ceda479d15bb7fe80ba761d5f23aa12467bd7558b43533a43fb2db01","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-08T22:02:28Z","title_canon_sha256":"5281d8d42ff155f0568c4150d75c210db8ee8d9a9b85939f1fbaee841c69cf34"},"schema_version":"1.0","source":{"id":"1906.03532","kind":"arxiv","version":2}},"canonical_sha256":"f924c2212c211b9e587f3b4831cd9fa10605efe3f8351d535b5220a6b2f080d7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f924c2212c211b9e587f3b4831cd9fa10605efe3f8351d535b5220a6b2f080d7","first_computed_at":"2026-05-17T23:42:59.867262Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:42:59.867262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7vMVZa+9VtMJVKb29YQ+W3Vqm40C+1Pkcac4KM25hDR9g4bSbVOi9chCtgLfi5JWHziK3WOQi9gHW2ri350EDA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:42:59.867690Z","signed_message":"canonical_sha256_bytes"},"source_id":"1906.03532","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:82e2c65a5475635dc02592d414ceefad69c99e2e07d2e176748513f79788a5a8","sha256:4f7c75f09448dcc7e9f0d417ebbd3b254ffd2ce5bca0ca4309e8fa7bf06501f3"],"state_sha256":"315556d8e4ee43c5c42f176ce95429ef591a5822bc9e72d05cb71d96d6040d53"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FdBC6ZucvHp5sv2Q3MWhrOP32fzoVfr2i1O9xXUzgDpe9MuC2krDdTPqynpeaoEXiai0p70Ffr+TDV/ihyzdBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T06:54:53.667067Z","bundle_sha256":"124ea3e5db3dd77ad9fa9c3d7df28d5a119a8ca9a312eda25767dff8815d6a84"}}