{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:7NHSIOCMI2LWNXLRAVAO3CWQW5","short_pith_number":"pith:7NHSIOCM","canonical_record":{"source":{"id":"1806.06763","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-18T15:17:01Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"b0a16ff62cca94fbe3b184fa6755a8df2e58de77dae0b0725bb0a483e7bc96fd","abstract_canon_sha256":"30447ac4c5e1d033c9c8d04eb287eafbff382e93a6e1d073b3b266774d940588"},"schema_version":"1.0"},"canonical_sha256":"fb4f24384c469766dd710540ed8ad0b7589dc771b4ef9fb9af011f5a9a3dc786","source":{"kind":"arxiv","id":"1806.06763","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.06763","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"arxiv_version","alias_value":"1806.06763v3","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.06763","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"pith_short_12","alias_value":"7NHSIOCMI2LW","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"pith_short_16","alias_value":"7NHSIOCMI2LWNXLR","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"pith_short_8","alias_value":"7NHSIOCM","created_at":"2026-07-05T01:12:16Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:7NHSIOCMI2LWNXLRAVAO3CWQW5","target":"record","payload":{"canonical_record":{"source":{"id":"1806.06763","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-18T15:17:01Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"b0a16ff62cca94fbe3b184fa6755a8df2e58de77dae0b0725bb0a483e7bc96fd","abstract_canon_sha256":"30447ac4c5e1d033c9c8d04eb287eafbff382e93a6e1d073b3b266774d940588"},"schema_version":"1.0"},"canonical_sha256":"fb4f24384c469766dd710540ed8ad0b7589dc771b4ef9fb9af011f5a9a3dc786","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:12:16.501385Z","signature_b64":"+IsFxqnRdMFePRLn8CWDfzY0tVSCK5Y7T1r2900+iB+Lk83aA+lVUUwhjTVjjB9UHEWX2aVCOWpFHMgeUZPRBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb4f24384c469766dd710540ed8ad0b7589dc771b4ef9fb9af011f5a9a3dc786","last_reissued_at":"2026-07-05T01:12:16.500913Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:12:16.500913Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1806.06763","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":"2026-07-05T01:12:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RIroeNyHomBqBT63jzKghu0dmX8O6h1lRE6bq+hjGQyCldh/FDbULCFY2pStk5uh/rzmW7RPuLWKPLPoKNfUBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:18:26.614878Z"},"content_sha256":"d5170e373b393bdd150fb82ca682983d7e4ddbde592350a8ff616f43f4c7730e","schema_version":"1.0","event_id":"sha256:d5170e373b393bdd150fb82ca682983d7e4ddbde592350a8ff616f43f4c7730e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:7NHSIOCMI2LWNXLRAVAO3CWQW5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Dongruo Zhou, Jinghui Chen, Quanquan Gu, Yiqi Tang, Yuan Cao, Ziyan Yang","submitted_at":"2018-06-18T15:17:01Z","abstract_excerpt":"Adaptive gradient methods, which adopt historical gradient information to automatically adjust the learning rate, despite the nice property of fast convergence, have been observed to generalize worse than stochastic gradient descent (SGD) with momentum in training deep neural networks. This leaves how to close the generalization gap of adaptive gradient methods an open problem. In this work, we show that adaptive gradient methods such as Adam, Amsgrad, are sometimes \"over adapted\". We design a new algorithm, called Partially adaptive momentum estimation method, which unifies the Adam/Amsgrad w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.06763","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1806.06763/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2026-07-05T01:12:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S1/CwYC3d2s2JUMN2YeddjrAXYPqgoOstlfx5E1S7ho7DwyvLfdEGvnj0bTQ8gAUUWUxERH4+qPknwDovtWVAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T13:18:26.615431Z"},"content_sha256":"1dc273432da74c0c95136b163d331f99acd0d172badc5f2ce7eab2170cd1f10d","schema_version":"1.0","event_id":"sha256:1dc273432da74c0c95136b163d331f99acd0d172badc5f2ce7eab2170cd1f10d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7NHSIOCMI2LWNXLRAVAO3CWQW5/bundle.json","state_url":"https://pith.science/pith/7NHSIOCMI2LWNXLRAVAO3CWQW5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7NHSIOCMI2LWNXLRAVAO3CWQW5/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":"2026-08-05T13:18:26Z","links":{"resolver":"https://pith.science/pith/7NHSIOCMI2LWNXLRAVAO3CWQW5","bundle":"https://pith.science/pith/7NHSIOCMI2LWNXLRAVAO3CWQW5/bundle.json","state":"https://pith.science/pith/7NHSIOCMI2LWNXLRAVAO3CWQW5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7NHSIOCMI2LWNXLRAVAO3CWQW5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:7NHSIOCMI2LWNXLRAVAO3CWQW5","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":"30447ac4c5e1d033c9c8d04eb287eafbff382e93a6e1d073b3b266774d940588","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-18T15:17:01Z","title_canon_sha256":"b0a16ff62cca94fbe3b184fa6755a8df2e58de77dae0b0725bb0a483e7bc96fd"},"schema_version":"1.0","source":{"id":"1806.06763","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.06763","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"arxiv_version","alias_value":"1806.06763v3","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.06763","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"pith_short_12","alias_value":"7NHSIOCMI2LW","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"pith_short_16","alias_value":"7NHSIOCMI2LWNXLR","created_at":"2026-07-05T01:12:16Z"},{"alias_kind":"pith_short_8","alias_value":"7NHSIOCM","created_at":"2026-07-05T01:12:16Z"}],"graph_snapshots":[{"event_id":"sha256:1dc273432da74c0c95136b163d331f99acd0d172badc5f2ce7eab2170cd1f10d","target":"graph","created_at":"2026-07-05T01:12:16Z","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/1806.06763/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Adaptive gradient methods, which adopt historical gradient information to automatically adjust the learning rate, despite the nice property of fast convergence, have been observed to generalize worse than stochastic gradient descent (SGD) with momentum in training deep neural networks. This leaves how to close the generalization gap of adaptive gradient methods an open problem. In this work, we show that adaptive gradient methods such as Adam, Amsgrad, are sometimes \"over adapted\". We design a new algorithm, called Partially adaptive momentum estimation method, which unifies the Adam/Amsgrad w","authors_text":"Dongruo Zhou, Jinghui Chen, Quanquan Gu, Yiqi Tang, Yuan Cao, Ziyan Yang","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-18T15:17:01Z","title":"Closing the Generalization Gap of Adaptive Gradient Methods in Training Deep Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.06763","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:d5170e373b393bdd150fb82ca682983d7e4ddbde592350a8ff616f43f4c7730e","target":"record","created_at":"2026-07-05T01:12:16Z","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":"30447ac4c5e1d033c9c8d04eb287eafbff382e93a6e1d073b3b266774d940588","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-18T15:17:01Z","title_canon_sha256":"b0a16ff62cca94fbe3b184fa6755a8df2e58de77dae0b0725bb0a483e7bc96fd"},"schema_version":"1.0","source":{"id":"1806.06763","kind":"arxiv","version":3}},"canonical_sha256":"fb4f24384c469766dd710540ed8ad0b7589dc771b4ef9fb9af011f5a9a3dc786","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fb4f24384c469766dd710540ed8ad0b7589dc771b4ef9fb9af011f5a9a3dc786","first_computed_at":"2026-07-05T01:12:16.500913Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:12:16.500913Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+IsFxqnRdMFePRLn8CWDfzY0tVSCK5Y7T1r2900+iB+Lk83aA+lVUUwhjTVjjB9UHEWX2aVCOWpFHMgeUZPRBw==","signature_status":"signed_v1","signed_at":"2026-07-05T01:12:16.501385Z","signed_message":"canonical_sha256_bytes"},"source_id":"1806.06763","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d5170e373b393bdd150fb82ca682983d7e4ddbde592350a8ff616f43f4c7730e","sha256:1dc273432da74c0c95136b163d331f99acd0d172badc5f2ce7eab2170cd1f10d"],"state_sha256":"0da9b6ef3a6aa3ba9a93043660294d9931fa2d9cd8619622b1b962a46b6f244c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QEY5oVOtgjM/OG7WjV/bUDWXjwNihPLHqPm4eh0f1599M2augWWPyPvBiHRINwy1KxMUjG8nbz8GoQhfHGs6Dg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T13:18:26.619613Z","bundle_sha256":"8297bbe78c6ba11e867a2e1bae1e2e28548766580148cd69fc387018e3d638cf"}}