{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:ZQ2NUWCT6KGSBSH6MVQBPRBXBC","short_pith_number":"pith:ZQ2NUWCT","canonical_record":{"source":{"id":"1807.06766","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-18T03:58:02Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"47e834fa51179e1ac8f8c1e096ecc9e9fd9de3306ebb4deb374ae0ce11ef2dc1","abstract_canon_sha256":"8ec46d9ebd4dc6ea613bf23db738245543d91c0be72a6c4a96249b715b6c44d2"},"schema_version":"1.0"},"canonical_sha256":"cc34da5853f28d20c8fe656017c43708bc59c3fd1c9be5ede377ad260e54b691","source":{"kind":"arxiv","id":"1807.06766","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.06766","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"arxiv_version","alias_value":"1807.06766v3","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.06766","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"pith_short_12","alias_value":"ZQ2NUWCT6KGS","created_at":"2026-05-18T12:33:07Z"},{"alias_kind":"pith_short_16","alias_value":"ZQ2NUWCT6KGSBSH6","created_at":"2026-05-18T12:33:07Z"},{"alias_kind":"pith_short_8","alias_value":"ZQ2NUWCT","created_at":"2026-05-18T12:33:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:ZQ2NUWCT6KGSBSH6MVQBPRBXBC","target":"record","payload":{"canonical_record":{"source":{"id":"1807.06766","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-18T03:58:02Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"47e834fa51179e1ac8f8c1e096ecc9e9fd9de3306ebb4deb374ae0ce11ef2dc1","abstract_canon_sha256":"8ec46d9ebd4dc6ea613bf23db738245543d91c0be72a6c4a96249b715b6c44d2"},"schema_version":"1.0"},"canonical_sha256":"cc34da5853f28d20c8fe656017c43708bc59c3fd1c9be5ede377ad260e54b691","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:00:13.014426Z","signature_b64":"8JO02zFXKL0N4tCyeLvn3400LDQzfdZl5eF5m4A+Xx9kVs3UtctrQEK3/M9CDsMgeY/iSuO1a8b0skxJeGPtCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cc34da5853f28d20c8fe656017c43708bc59c3fd1c9be5ede377ad260e54b691","last_reissued_at":"2026-05-18T00:00:13.013864Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:00:13.013864Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1807.06766","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:00:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zH1D+Jo5GgCC+VUL6bSUcYA1gAedhUUm1WF3Ytb2GPuRcrX5lJoHdE+bdf68ia8NEaFfbuVl8Rf1V9079RZ4DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-20T02:43:47.844488Z"},"content_sha256":"186049586f7bc5bc6f6c2ddc9e1fa21af0458f8cb4c2e4269beb2493677bec33","schema_version":"1.0","event_id":"sha256:186049586f7bc5bc6f6c2ddc9e1fa21af0458f8cb4c2e4269beb2493677bec33"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:ZQ2NUWCT6KGSBSH6MVQBPRBXBC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Convergence guarantees for RMSProp and ADAM in non-convex optimization and an empirical comparison to Nesterov acceleration","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anirbit Mukherjee, Enayat Ullah, Soham De","submitted_at":"2018-07-18T03:58:02Z","abstract_excerpt":"RMSProp and ADAM continue to be extremely popular algorithms for training neural nets but their theoretical convergence properties have remained unclear. Further, recent work has seemed to suggest that these algorithms have worse generalization properties when compared to carefully tuned stochastic gradient descent or its momentum variants. In this work, we make progress towards a deeper understanding of ADAM and RMSProp in two ways. First, we provide proofs that these adaptive gradient algorithms are guaranteed to reach criticality for smooth non-convex objectives, and we give bounds on the r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.06766","kind":"arxiv","version":3},"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":"2026-05-18T00:00:13Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BFpyJX6DgPIQqtwSRMT0GPXc3GlGtQDczK5sY/IqdH07y4PLWGPKQgERaxn5VsjoUM1vtHO78TDLmCsVnklADw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-20T02:43:47.844835Z"},"content_sha256":"7dc94e18c2ebbdc4bbde758cd69cf45319c560de61dc11bb06ca6770e415f458","schema_version":"1.0","event_id":"sha256:7dc94e18c2ebbdc4bbde758cd69cf45319c560de61dc11bb06ca6770e415f458"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZQ2NUWCT6KGSBSH6MVQBPRBXBC/bundle.json","state_url":"https://pith.science/pith/ZQ2NUWCT6KGSBSH6MVQBPRBXBC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZQ2NUWCT6KGSBSH6MVQBPRBXBC/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":"2026-06-20T02:43:47Z","links":{"resolver":"https://pith.science/pith/ZQ2NUWCT6KGSBSH6MVQBPRBXBC","bundle":"https://pith.science/pith/ZQ2NUWCT6KGSBSH6MVQBPRBXBC/bundle.json","state":"https://pith.science/pith/ZQ2NUWCT6KGSBSH6MVQBPRBXBC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZQ2NUWCT6KGSBSH6MVQBPRBXBC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:ZQ2NUWCT6KGSBSH6MVQBPRBXBC","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":"8ec46d9ebd4dc6ea613bf23db738245543d91c0be72a6c4a96249b715b6c44d2","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-18T03:58:02Z","title_canon_sha256":"47e834fa51179e1ac8f8c1e096ecc9e9fd9de3306ebb4deb374ae0ce11ef2dc1"},"schema_version":"1.0","source":{"id":"1807.06766","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.06766","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"arxiv_version","alias_value":"1807.06766v3","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.06766","created_at":"2026-05-18T00:00:13Z"},{"alias_kind":"pith_short_12","alias_value":"ZQ2NUWCT6KGS","created_at":"2026-05-18T12:33:07Z"},{"alias_kind":"pith_short_16","alias_value":"ZQ2NUWCT6KGSBSH6","created_at":"2026-05-18T12:33:07Z"},{"alias_kind":"pith_short_8","alias_value":"ZQ2NUWCT","created_at":"2026-05-18T12:33:07Z"}],"graph_snapshots":[{"event_id":"sha256:7dc94e18c2ebbdc4bbde758cd69cf45319c560de61dc11bb06ca6770e415f458","target":"graph","created_at":"2026-05-18T00:00:13Z","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":"RMSProp and ADAM continue to be extremely popular algorithms for training neural nets but their theoretical convergence properties have remained unclear. Further, recent work has seemed to suggest that these algorithms have worse generalization properties when compared to carefully tuned stochastic gradient descent or its momentum variants. In this work, we make progress towards a deeper understanding of ADAM and RMSProp in two ways. First, we provide proofs that these adaptive gradient algorithms are guaranteed to reach criticality for smooth non-convex objectives, and we give bounds on the r","authors_text":"Anirbit Mukherjee, Enayat Ullah, Soham De","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-18T03:58:02Z","title":"Convergence guarantees for RMSProp and ADAM in non-convex optimization and an empirical comparison to Nesterov acceleration"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.06766","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":"sha256:186049586f7bc5bc6f6c2ddc9e1fa21af0458f8cb4c2e4269beb2493677bec33","target":"record","created_at":"2026-05-18T00:00:13Z","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":"8ec46d9ebd4dc6ea613bf23db738245543d91c0be72a6c4a96249b715b6c44d2","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-18T03:58:02Z","title_canon_sha256":"47e834fa51179e1ac8f8c1e096ecc9e9fd9de3306ebb4deb374ae0ce11ef2dc1"},"schema_version":"1.0","source":{"id":"1807.06766","kind":"arxiv","version":3}},"canonical_sha256":"cc34da5853f28d20c8fe656017c43708bc59c3fd1c9be5ede377ad260e54b691","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cc34da5853f28d20c8fe656017c43708bc59c3fd1c9be5ede377ad260e54b691","first_computed_at":"2026-05-18T00:00:13.013864Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:00:13.013864Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8JO02zFXKL0N4tCyeLvn3400LDQzfdZl5eF5m4A+Xx9kVs3UtctrQEK3/M9CDsMgeY/iSuO1a8b0skxJeGPtCw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:00:13.014426Z","signed_message":"canonical_sha256_bytes"},"source_id":"1807.06766","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:186049586f7bc5bc6f6c2ddc9e1fa21af0458f8cb4c2e4269beb2493677bec33","sha256:7dc94e18c2ebbdc4bbde758cd69cf45319c560de61dc11bb06ca6770e415f458"],"state_sha256":"6359e738b0d2aa14cfff595fe8684166655b34bbf58317f7862430565f17fe04"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xR93CJjpTy9DAj+j422xYhWyPCNnlrpm3uAaSDI6L3Oa3bC8N7BoTc0QPiiX3ZCASeONN4fD/MzfYldvbgWzAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-20T02:43:47.846707Z","bundle_sha256":"4fcfb32fbe55fc04b62f74a7958fcb2b5975beeeae24eadea5521ed13630b0e4"}}