{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:CWPYLZDW7AWAX7TQ5ZRNS3OHEM","short_pith_number":"pith:CWPYLZDW","canonical_record":{"source":{"id":"1805.07898","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T05:28:22Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"2ceefe1c53ad3f1df78e1d1b862b5721652f4ce3091f306c744843430658788f","abstract_canon_sha256":"6fb0cb15c960c74a714fea5e049b367e02246d479a6cb7039005eed483ca73d7"},"schema_version":"1.0"},"canonical_sha256":"159f85e476f82c0bfe70ee62d96dc723359589aaf4c6603cb3ca31edfe59916c","source":{"kind":"arxiv","id":"1805.07898","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.07898","created_at":"2026-05-17T23:59:22Z"},{"alias_kind":"arxiv_version","alias_value":"1805.07898v3","created_at":"2026-05-17T23:59:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.07898","created_at":"2026-05-17T23:59:22Z"},{"alias_kind":"pith_short_12","alias_value":"CWPYLZDW7AWA","created_at":"2026-05-18T12:32:19Z"},{"alias_kind":"pith_short_16","alias_value":"CWPYLZDW7AWAX7TQ","created_at":"2026-05-18T12:32:19Z"},{"alias_kind":"pith_short_8","alias_value":"CWPYLZDW","created_at":"2026-05-18T12:32:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:CWPYLZDW7AWAX7TQ5ZRNS3OHEM","target":"record","payload":{"canonical_record":{"source":{"id":"1805.07898","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T05:28:22Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"2ceefe1c53ad3f1df78e1d1b862b5721652f4ce3091f306c744843430658788f","abstract_canon_sha256":"6fb0cb15c960c74a714fea5e049b367e02246d479a6cb7039005eed483ca73d7"},"schema_version":"1.0"},"canonical_sha256":"159f85e476f82c0bfe70ee62d96dc723359589aaf4c6603cb3ca31edfe59916c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:22.698636Z","signature_b64":"WxdwGuXLdzeT4hcVQ8AmxTy/X5n6M+YzW7f6GyqnIAa8XF6wJISRvQHoLojSzzOOlZS4s8AdJGwHblzfM9TwBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"159f85e476f82c0bfe70ee62d96dc723359589aaf4c6603cb3ca31edfe59916c","last_reissued_at":"2026-05-17T23:59:22.698266Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:22.698266Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1805.07898","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":"2026-05-17T23:59:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IeRPImsmH/wBNq4t35kVuBhg5K7O0YmptCLDkumc/wPpOI7LudndEFqGTKJ4kB/X9du1vWkoIgGurx6UJH19Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T04:00:15.810250Z"},"content_sha256":"34a28022ca3d6a13ed3d6ad8bb8721d9eead05bcdeb5c8c17e86d8e86b5676a6","schema_version":"1.0","event_id":"sha256:34a28022ca3d6a13ed3d6ad8bb8721d9eead05bcdeb5c8c17e86d8e86b5676a6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:CWPYLZDW7AWAX7TQ5ZRNS3OHEM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SmoothOut: Smoothing Out Sharp Minima to Improve Generalization in Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Chunpeng Wu, Cong Xu, Feng Yan, Hai Li, Wei Wen, Yandan Wang, Yiran Chen","submitted_at":"2018-05-21T05:28:22Z","abstract_excerpt":"In Deep Learning, Stochastic Gradient Descent (SGD) is usually selected as a training method because of its efficiency; however, recently, a problem in SGD gains research interest: sharp minima in Deep Neural Networks (DNNs) have poor generalization; especially, large-batch SGD tends to converge to sharp minima. It becomes an open question whether escaping sharp minima can improve the generalization. To answer this question, we propose SmoothOut framework to smooth out sharp minima in DNNs and thereby improve generalization. In a nutshell, SmoothOut perturbs multiple copies of the DNN by noise"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.07898","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":"2026-05-17T23:59:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lzTQsICbsu1AdLPEwieTRHGTaRFakgjpX+h085ons3rADjoHbP91+McHlhYgmOfiPW3iqDPlDuMDjsqwX5YoBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T04:00:15.811323Z"},"content_sha256":"4cc9344b7ef09146e3f7e3499ce80d9d1d589e3d1b1e7405bccd202191772a59","schema_version":"1.0","event_id":"sha256:4cc9344b7ef09146e3f7e3499ce80d9d1d589e3d1b1e7405bccd202191772a59"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CWPYLZDW7AWAX7TQ5ZRNS3OHEM/bundle.json","state_url":"https://pith.science/pith/CWPYLZDW7AWAX7TQ5ZRNS3OHEM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CWPYLZDW7AWAX7TQ5ZRNS3OHEM/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":"2026-05-26T04:00:15Z","links":{"resolver":"https://pith.science/pith/CWPYLZDW7AWAX7TQ5ZRNS3OHEM","bundle":"https://pith.science/pith/CWPYLZDW7AWAX7TQ5ZRNS3OHEM/bundle.json","state":"https://pith.science/pith/CWPYLZDW7AWAX7TQ5ZRNS3OHEM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CWPYLZDW7AWAX7TQ5ZRNS3OHEM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:CWPYLZDW7AWAX7TQ5ZRNS3OHEM","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":"6fb0cb15c960c74a714fea5e049b367e02246d479a6cb7039005eed483ca73d7","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T05:28:22Z","title_canon_sha256":"2ceefe1c53ad3f1df78e1d1b862b5721652f4ce3091f306c744843430658788f"},"schema_version":"1.0","source":{"id":"1805.07898","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.07898","created_at":"2026-05-17T23:59:22Z"},{"alias_kind":"arxiv_version","alias_value":"1805.07898v3","created_at":"2026-05-17T23:59:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.07898","created_at":"2026-05-17T23:59:22Z"},{"alias_kind":"pith_short_12","alias_value":"CWPYLZDW7AWA","created_at":"2026-05-18T12:32:19Z"},{"alias_kind":"pith_short_16","alias_value":"CWPYLZDW7AWAX7TQ","created_at":"2026-05-18T12:32:19Z"},{"alias_kind":"pith_short_8","alias_value":"CWPYLZDW","created_at":"2026-05-18T12:32:19Z"}],"graph_snapshots":[{"event_id":"sha256:4cc9344b7ef09146e3f7e3499ce80d9d1d589e3d1b1e7405bccd202191772a59","target":"graph","created_at":"2026-05-17T23:59:22Z","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":"In Deep Learning, Stochastic Gradient Descent (SGD) is usually selected as a training method because of its efficiency; however, recently, a problem in SGD gains research interest: sharp minima in Deep Neural Networks (DNNs) have poor generalization; especially, large-batch SGD tends to converge to sharp minima. It becomes an open question whether escaping sharp minima can improve the generalization. To answer this question, we propose SmoothOut framework to smooth out sharp minima in DNNs and thereby improve generalization. In a nutshell, SmoothOut perturbs multiple copies of the DNN by noise","authors_text":"Chunpeng Wu, Cong Xu, Feng Yan, Hai Li, Wei Wen, Yandan Wang, Yiran Chen","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T05:28:22Z","title":"SmoothOut: Smoothing Out Sharp Minima to Improve Generalization in Deep Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.07898","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":"sha256:34a28022ca3d6a13ed3d6ad8bb8721d9eead05bcdeb5c8c17e86d8e86b5676a6","target":"record","created_at":"2026-05-17T23:59:22Z","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":"6fb0cb15c960c74a714fea5e049b367e02246d479a6cb7039005eed483ca73d7","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-05-21T05:28:22Z","title_canon_sha256":"2ceefe1c53ad3f1df78e1d1b862b5721652f4ce3091f306c744843430658788f"},"schema_version":"1.0","source":{"id":"1805.07898","kind":"arxiv","version":3}},"canonical_sha256":"159f85e476f82c0bfe70ee62d96dc723359589aaf4c6603cb3ca31edfe59916c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"159f85e476f82c0bfe70ee62d96dc723359589aaf4c6603cb3ca31edfe59916c","first_computed_at":"2026-05-17T23:59:22.698266Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:59:22.698266Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WxdwGuXLdzeT4hcVQ8AmxTy/X5n6M+YzW7f6GyqnIAa8XF6wJISRvQHoLojSzzOOlZS4s8AdJGwHblzfM9TwBw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:59:22.698636Z","signed_message":"canonical_sha256_bytes"},"source_id":"1805.07898","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:34a28022ca3d6a13ed3d6ad8bb8721d9eead05bcdeb5c8c17e86d8e86b5676a6","sha256:4cc9344b7ef09146e3f7e3499ce80d9d1d589e3d1b1e7405bccd202191772a59"],"state_sha256":"fd5c4b2efae9f9d6389411b12acbd3df2794a2d6b050e825113795b9381cf281"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tXtw5quzXQDu97VKUBu1EUljLn2RZjBlv77JqG1orXg2AXwPBs2RqBt94CXFSDIpWNpyG3RXxhdhtYgIKsutAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T04:00:15.814592Z","bundle_sha256":"e534e2cd95a88ad20715837f202c27c5517e52f63dd135eb13ebfd77b4c97718"}}