{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:SETOAZEYAMXLS2ZIS56Z4PQUM4","short_pith_number":"pith:SETOAZEY","canonical_record":{"source":{"id":"1803.05407","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-14T17:09:27Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"d607ccfd50cde3bb43b1255345618b2b00a1b5046f1516107637080d707abb64","abstract_canon_sha256":"5e265993e34fccbbb2148419991c16061c9177440d5fb622841a059d5e441e7d"},"schema_version":"1.0"},"canonical_sha256":"9126e06498032eb96b28977d9e3e146721a5a4ccda077940e46273ad7ec17c3b","source":{"kind":"arxiv","id":"1803.05407","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1803.05407","created_at":"2026-05-17T23:52:53Z"},{"alias_kind":"arxiv_version","alias_value":"1803.05407v3","created_at":"2026-05-17T23:52:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.05407","created_at":"2026-05-17T23:52:53Z"},{"alias_kind":"pith_short_12","alias_value":"SETOAZEYAMXL","created_at":"2026-05-18T12:32:53Z"},{"alias_kind":"pith_short_16","alias_value":"SETOAZEYAMXLS2ZI","created_at":"2026-05-18T12:32:53Z"},{"alias_kind":"pith_short_8","alias_value":"SETOAZEY","created_at":"2026-05-18T12:32:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:SETOAZEYAMXLS2ZIS56Z4PQUM4","target":"record","payload":{"canonical_record":{"source":{"id":"1803.05407","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-14T17:09:27Z","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"title_canon_sha256":"d607ccfd50cde3bb43b1255345618b2b00a1b5046f1516107637080d707abb64","abstract_canon_sha256":"5e265993e34fccbbb2148419991c16061c9177440d5fb622841a059d5e441e7d"},"schema_version":"1.0"},"canonical_sha256":"9126e06498032eb96b28977d9e3e146721a5a4ccda077940e46273ad7ec17c3b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:52:53.272097Z","signature_b64":"lTS72qpnAt9JA3LEe+BLu0X8Aqwr1AdXeV92JQcNIHq/ajJqUicU2G3kmGG6vzk8CXju6TVFdk/mghXZ/+jUAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9126e06498032eb96b28977d9e3e146721a5a4ccda077940e46273ad7ec17c3b","last_reissued_at":"2026-05-17T23:52:53.271272Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:52:53.271272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1803.05407","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":"2026-05-17T23:52:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"35syEDs+eP544L1BXB3FDoqVfN4qOH184otTdWGlcIwh8+DadS6c090GA2s9/xqqZguwObYY++rZeirT+Kn4Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T21:03:26.638353Z"},"content_sha256":"d38cf344dc454386eb135038e5b4795a9374ed256c8d4eb8b6b91b64c5262910","schema_version":"1.0","event_id":"sha256:d38cf344dc454386eb135038e5b4795a9374ed256c8d4eb8b6b91b64c5262910"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:SETOAZEYAMXLS2ZIS56Z4PQUM4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Averaging Weights Leads to Wider Optima and Better Generalization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Gordon Wilson, Dmitrii Podoprikhin, Dmitry Vetrov, Pavel Izmailov, Timur Garipov","submitted_at":"2018-03-14T17:09:27Z","abstract_excerpt":"Deep neural networks are typically trained by optimizing a loss function with an SGD variant, in conjunction with a decaying learning rate, until convergence. We show that simple averaging of multiple points along the trajectory of SGD, with a cyclical or constant learning rate, leads to better generalization than conventional training. We also show that this Stochastic Weight Averaging (SWA) procedure finds much flatter solutions than SGD, and approximates the recent Fast Geometric Ensembling (FGE) approach with a single model. Using SWA we achieve notable improvement in test accuracy over co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.05407","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":"2026-05-17T23:52:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fJObvDBVYASoETxOT8kOlRR09A6valoFXJqmMjjq6tHdhv8mU+tcA5CjGT52Ec8IApAezLiTy4j2+X7CVG0xAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-25T21:03:26.639012Z"},"content_sha256":"1fd2528dc6b4f30c95c7ee5622c1d81c81a864048445da3b99f958b00fe8e14d","schema_version":"1.0","event_id":"sha256:1fd2528dc6b4f30c95c7ee5622c1d81c81a864048445da3b99f958b00fe8e14d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SETOAZEYAMXLS2ZIS56Z4PQUM4/bundle.json","state_url":"https://pith.science/pith/SETOAZEYAMXLS2ZIS56Z4PQUM4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SETOAZEYAMXLS2ZIS56Z4PQUM4/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":"2026-05-25T21:03:26Z","links":{"resolver":"https://pith.science/pith/SETOAZEYAMXLS2ZIS56Z4PQUM4","bundle":"https://pith.science/pith/SETOAZEYAMXLS2ZIS56Z4PQUM4/bundle.json","state":"https://pith.science/pith/SETOAZEYAMXLS2ZIS56Z4PQUM4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SETOAZEYAMXLS2ZIS56Z4PQUM4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:SETOAZEYAMXLS2ZIS56Z4PQUM4","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":"5e265993e34fccbbb2148419991c16061c9177440d5fb622841a059d5e441e7d","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-14T17:09:27Z","title_canon_sha256":"d607ccfd50cde3bb43b1255345618b2b00a1b5046f1516107637080d707abb64"},"schema_version":"1.0","source":{"id":"1803.05407","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1803.05407","created_at":"2026-05-17T23:52:53Z"},{"alias_kind":"arxiv_version","alias_value":"1803.05407v3","created_at":"2026-05-17T23:52:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1803.05407","created_at":"2026-05-17T23:52:53Z"},{"alias_kind":"pith_short_12","alias_value":"SETOAZEYAMXL","created_at":"2026-05-18T12:32:53Z"},{"alias_kind":"pith_short_16","alias_value":"SETOAZEYAMXLS2ZI","created_at":"2026-05-18T12:32:53Z"},{"alias_kind":"pith_short_8","alias_value":"SETOAZEY","created_at":"2026-05-18T12:32:53Z"}],"graph_snapshots":[{"event_id":"sha256:1fd2528dc6b4f30c95c7ee5622c1d81c81a864048445da3b99f958b00fe8e14d","target":"graph","created_at":"2026-05-17T23:52:53Z","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":"Deep neural networks are typically trained by optimizing a loss function with an SGD variant, in conjunction with a decaying learning rate, until convergence. We show that simple averaging of multiple points along the trajectory of SGD, with a cyclical or constant learning rate, leads to better generalization than conventional training. We also show that this Stochastic Weight Averaging (SWA) procedure finds much flatter solutions than SGD, and approximates the recent Fast Geometric Ensembling (FGE) approach with a single model. Using SWA we achieve notable improvement in test accuracy over co","authors_text":"Andrew Gordon Wilson, Dmitrii Podoprikhin, Dmitry Vetrov, Pavel Izmailov, Timur Garipov","cross_cats":["cs.AI","cs.CV","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-14T17:09:27Z","title":"Averaging Weights Leads to Wider Optima and Better Generalization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1803.05407","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:d38cf344dc454386eb135038e5b4795a9374ed256c8d4eb8b6b91b64c5262910","target":"record","created_at":"2026-05-17T23:52:53Z","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":"5e265993e34fccbbb2148419991c16061c9177440d5fb622841a059d5e441e7d","cross_cats_sorted":["cs.AI","cs.CV","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-03-14T17:09:27Z","title_canon_sha256":"d607ccfd50cde3bb43b1255345618b2b00a1b5046f1516107637080d707abb64"},"schema_version":"1.0","source":{"id":"1803.05407","kind":"arxiv","version":3}},"canonical_sha256":"9126e06498032eb96b28977d9e3e146721a5a4ccda077940e46273ad7ec17c3b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9126e06498032eb96b28977d9e3e146721a5a4ccda077940e46273ad7ec17c3b","first_computed_at":"2026-05-17T23:52:53.271272Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:52:53.271272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lTS72qpnAt9JA3LEe+BLu0X8Aqwr1AdXeV92JQcNIHq/ajJqUicU2G3kmGG6vzk8CXju6TVFdk/mghXZ/+jUAA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:52:53.272097Z","signed_message":"canonical_sha256_bytes"},"source_id":"1803.05407","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d38cf344dc454386eb135038e5b4795a9374ed256c8d4eb8b6b91b64c5262910","sha256:1fd2528dc6b4f30c95c7ee5622c1d81c81a864048445da3b99f958b00fe8e14d"],"state_sha256":"74053dffd29fe0886e10a1df0a3322af64b1093531a6ab9bdaae54bb32dcf2c7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Hbq1OHReWeAbbbcmF7yNGDr2G9Z2jLtxRin+Yw9+Aei/x5s8UvIt96OKM6sZrHPrrRNKhYfzuHWhhvLlt762DA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-25T21:03:26.642882Z","bundle_sha256":"0afbc1608982f5f3fd984928075483f029700dde8b07242f92ebbf28cefd64db"}}