{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:UXNRSXOAIEWSDAZYC4JGHKZRPT","short_pith_number":"pith:UXNRSXOA","canonical_record":{"source":{"id":"1807.08140","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-21T12:32:36Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"cb6881f5acf1e7b8aaebc1f50ce42e22d245e3a49a022c8a99e715392debfe21","abstract_canon_sha256":"0e511cfcba0d07dfa6da89e95adfa67123120062ddb394cdcd9c6dd9129d587e"},"schema_version":"1.0"},"canonical_sha256":"a5db195dc0412d218338171263ab317cc4fbb2163f0218c4953284aa751fcf4a","source":{"kind":"arxiv","id":"1807.08140","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.08140","created_at":"2026-05-18T00:10:09Z"},{"alias_kind":"arxiv_version","alias_value":"1807.08140v1","created_at":"2026-05-18T00:10:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.08140","created_at":"2026-05-18T00:10:09Z"},{"alias_kind":"pith_short_12","alias_value":"UXNRSXOAIEWS","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"UXNRSXOAIEWSDAZY","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"UXNRSXOA","created_at":"2026-05-18T12:32:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:UXNRSXOAIEWSDAZYC4JGHKZRPT","target":"record","payload":{"canonical_record":{"source":{"id":"1807.08140","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-21T12:32:36Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"cb6881f5acf1e7b8aaebc1f50ce42e22d245e3a49a022c8a99e715392debfe21","abstract_canon_sha256":"0e511cfcba0d07dfa6da89e95adfa67123120062ddb394cdcd9c6dd9129d587e"},"schema_version":"1.0"},"canonical_sha256":"a5db195dc0412d218338171263ab317cc4fbb2163f0218c4953284aa751fcf4a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:10:09.151375Z","signature_b64":"T7zY4fOtzI6LBEm6GiixU5J6+gtSa04m3T9pPfMrDsB5D1ntkfoALz+rg07WdCSVfFMS7lUTN6zQ3b8k4PkwDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5db195dc0412d218338171263ab317cc4fbb2163f0218c4953284aa751fcf4a","last_reissued_at":"2026-05-18T00:10:09.150817Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:10:09.150817Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1807.08140","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:10:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tN8t1eJZZwobwUHedmEePpDoHSZzN8GdkADsD7J2YuQswKOyu2dcXXTuyVXhq24ojN17weKDXF4pgZ4sY6jRBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T03:06:07.094567Z"},"content_sha256":"5917824b9ab8f825cfab100abaf12aa40d64d2040f7376a4d986fbd80f204cd8","schema_version":"1.0","event_id":"sha256:5917824b9ab8f825cfab100abaf12aa40d64d2040f7376a4d986fbd80f204cd8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:UXNRSXOAIEWSDAZYC4JGHKZRPT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Analysis of Trajectories of Gradient Descent in the Optimization of Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Adepu Ravi Sankar, Vineeth N Balasubramanian, Vishwak Srinivasan","submitted_at":"2018-07-21T12:32:36Z","abstract_excerpt":"Theoretical analysis of the error landscape of deep neural networks has garnered significant interest in recent years. In this work, we theoretically study the importance of noise in the trajectories of gradient descent towards optimal solutions in multi-layer neural networks. We show that adding noise (in different ways) to a neural network while training increases the rank of the product of weight matrices of a multi-layer linear neural network. We thus study how adding noise can assist reaching a global optimum when the product matrix is full-rank (under certain conditions). We establish th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.08140","kind":"arxiv","version":1},"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":"2026-05-18T00:10:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h0RrseHJrXpU9e1IOSv7Wc9whj0hRnNez0YLesOtOHgjQLG3nMLUj1+UmU+MFvaYJ2j0aTPxvemUgk+dE+FUDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T03:06:07.094923Z"},"content_sha256":"7fb4290abcc524ab1704ae092ff14ac13981624f5e21940c9235177bd530f5d2","schema_version":"1.0","event_id":"sha256:7fb4290abcc524ab1704ae092ff14ac13981624f5e21940c9235177bd530f5d2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UXNRSXOAIEWSDAZYC4JGHKZRPT/bundle.json","state_url":"https://pith.science/pith/UXNRSXOAIEWSDAZYC4JGHKZRPT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UXNRSXOAIEWSDAZYC4JGHKZRPT/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":"2026-07-04T03:06:07Z","links":{"resolver":"https://pith.science/pith/UXNRSXOAIEWSDAZYC4JGHKZRPT","bundle":"https://pith.science/pith/UXNRSXOAIEWSDAZYC4JGHKZRPT/bundle.json","state":"https://pith.science/pith/UXNRSXOAIEWSDAZYC4JGHKZRPT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UXNRSXOAIEWSDAZYC4JGHKZRPT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:UXNRSXOAIEWSDAZYC4JGHKZRPT","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":"0e511cfcba0d07dfa6da89e95adfa67123120062ddb394cdcd9c6dd9129d587e","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-21T12:32:36Z","title_canon_sha256":"cb6881f5acf1e7b8aaebc1f50ce42e22d245e3a49a022c8a99e715392debfe21"},"schema_version":"1.0","source":{"id":"1807.08140","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1807.08140","created_at":"2026-05-18T00:10:09Z"},{"alias_kind":"arxiv_version","alias_value":"1807.08140v1","created_at":"2026-05-18T00:10:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1807.08140","created_at":"2026-05-18T00:10:09Z"},{"alias_kind":"pith_short_12","alias_value":"UXNRSXOAIEWS","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_16","alias_value":"UXNRSXOAIEWSDAZY","created_at":"2026-05-18T12:32:56Z"},{"alias_kind":"pith_short_8","alias_value":"UXNRSXOA","created_at":"2026-05-18T12:32:56Z"}],"graph_snapshots":[{"event_id":"sha256:7fb4290abcc524ab1704ae092ff14ac13981624f5e21940c9235177bd530f5d2","target":"graph","created_at":"2026-05-18T00:10:09Z","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":"Theoretical analysis of the error landscape of deep neural networks has garnered significant interest in recent years. In this work, we theoretically study the importance of noise in the trajectories of gradient descent towards optimal solutions in multi-layer neural networks. We show that adding noise (in different ways) to a neural network while training increases the rank of the product of weight matrices of a multi-layer linear neural network. We thus study how adding noise can assist reaching a global optimum when the product matrix is full-rank (under certain conditions). We establish th","authors_text":"Adepu Ravi Sankar, Vineeth N Balasubramanian, Vishwak Srinivasan","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-21T12:32:36Z","title":"On the Analysis of Trajectories of Gradient Descent in the Optimization of Deep Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1807.08140","kind":"arxiv","version":1},"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":"sha256:5917824b9ab8f825cfab100abaf12aa40d64d2040f7376a4d986fbd80f204cd8","target":"record","created_at":"2026-05-18T00:10:09Z","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":"0e511cfcba0d07dfa6da89e95adfa67123120062ddb394cdcd9c6dd9129d587e","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-07-21T12:32:36Z","title_canon_sha256":"cb6881f5acf1e7b8aaebc1f50ce42e22d245e3a49a022c8a99e715392debfe21"},"schema_version":"1.0","source":{"id":"1807.08140","kind":"arxiv","version":1}},"canonical_sha256":"a5db195dc0412d218338171263ab317cc4fbb2163f0218c4953284aa751fcf4a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a5db195dc0412d218338171263ab317cc4fbb2163f0218c4953284aa751fcf4a","first_computed_at":"2026-05-18T00:10:09.150817Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:10:09.150817Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"T7zY4fOtzI6LBEm6GiixU5J6+gtSa04m3T9pPfMrDsB5D1ntkfoALz+rg07WdCSVfFMS7lUTN6zQ3b8k4PkwDA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:10:09.151375Z","signed_message":"canonical_sha256_bytes"},"source_id":"1807.08140","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5917824b9ab8f825cfab100abaf12aa40d64d2040f7376a4d986fbd80f204cd8","sha256:7fb4290abcc524ab1704ae092ff14ac13981624f5e21940c9235177bd530f5d2"],"state_sha256":"4801be6d4b5f3a7e864b0979f7cecf063e0b80f91c528f4e86e05ff142002631"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yzgLiHK9yTvATmYH32bHZsCH3ttJurBDv1ylnbSaKj9vnV1Y76Jq/E4iuQceErjgL1Xx1dfr2wdd1HgxwL54Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T03:06:07.097470Z","bundle_sha256":"271a51a9ca9cf3a2768ebe8e847930693a155f91404501c47b8697f6accbe13a"}}