{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2014:HOQJBI2H4BX76RFZPS5J2WJIGO","short_pith_number":"pith:HOQJBI2H","schema_version":"1.0","canonical_sha256":"3ba090a347e06fff44b97cba9d592833b2f7a403ccdaa4ec625c7e1b73dcbacb","source":{"kind":"arxiv","id":"1406.2572","version":1},"attestation_state":"computed","paper":{"title":"Identifying and attacking the saddle point problem in high-dimensional non-convex optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Caglar Gulcehre, Kyunghyun Cho, Razvan Pascanu, Surya Ganguli, Yann Dauphin, Yoshua Bengio","submitted_at":"2014-06-10T14:52:14Z","abstract_excerpt":"A central challenge to many fields of science and engineering involves minimizing non-convex error functions over continuous, high dimensional spaces. Gradient descent or quasi-Newton methods are almost ubiquitously used to perform such minimizations, and it is often thought that a main source of difficulty for these local methods to find the global minimum is the proliferation of local minima with much higher error than the global minimum. Here we argue, based on results from statistical physics, random matrix theory, neural network theory, and empirical evidence, that a deeper and more profo"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"1406.2572","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2014-06-10T14:52:14Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"086e7fec60472e3746f4e4b2db34bce8698e54f9c02ab5ae2595a7ddc01246b6","abstract_canon_sha256":"6553882606fa129a0985d43c18e8d174e494398865945dedc5c86feab611d96b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:50:01.400567Z","signature_b64":"GHZ1rudgQlA9Mra5zYA3iZKhEKP9f5C+ayCEQNT+Q9kUMOe/jm6iDgsw1tRrrCmX8jq/SyF2m+GEJRpvXIk9Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3ba090a347e06fff44b97cba9d592833b2f7a403ccdaa4ec625c7e1b73dcbacb","last_reissued_at":"2026-05-18T02:50:01.400092Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:50:01.400092Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Identifying and attacking the saddle point problem in high-dimensional non-convex optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Caglar Gulcehre, Kyunghyun Cho, Razvan Pascanu, Surya Ganguli, Yann Dauphin, Yoshua Bengio","submitted_at":"2014-06-10T14:52:14Z","abstract_excerpt":"A central challenge to many fields of science and engineering involves minimizing non-convex error functions over continuous, high dimensional spaces. Gradient descent or quasi-Newton methods are almost ubiquitously used to perform such minimizations, and it is often thought that a main source of difficulty for these local methods to find the global minimum is the proliferation of local minima with much higher error than the global minimum. Here we argue, based on results from statistical physics, random matrix theory, neural network theory, and empirical evidence, that a deeper and more profo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1406.2572","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"},"aliases":[{"alias_kind":"arxiv","alias_value":"1406.2572","created_at":"2026-05-18T02:50:01.400158+00:00"},{"alias_kind":"arxiv_version","alias_value":"1406.2572v1","created_at":"2026-05-18T02:50:01.400158+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1406.2572","created_at":"2026-05-18T02:50:01.400158+00:00"},{"alias_kind":"pith_short_12","alias_value":"HOQJBI2H4BX7","created_at":"2026-05-18T12:28:30.664211+00:00"},{"alias_kind":"pith_short_16","alias_value":"HOQJBI2H4BX76RFZ","created_at":"2026-05-18T12:28:30.664211+00:00"},{"alias_kind":"pith_short_8","alias_value":"HOQJBI2H","created_at":"2026-05-18T12:28:30.664211+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.02345","citing_title":"Low Rank Based Subspace Inference for the Laplace Approximation of Bayesian Neural Networks","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09331","citing_title":"Dimension-Free Saddle-Point Escape in Muon","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO","json":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO.json","graph_json":"https://pith.science/api/pith-number/HOQJBI2H4BX76RFZPS5J2WJIGO/graph.json","events_json":"https://pith.science/api/pith-number/HOQJBI2H4BX76RFZPS5J2WJIGO/events.json","paper":"https://pith.science/paper/HOQJBI2H"},"agent_actions":{"view_html":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO","download_json":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO.json","view_paper":"https://pith.science/paper/HOQJBI2H","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1406.2572&json=true","fetch_graph":"https://pith.science/api/pith-number/HOQJBI2H4BX76RFZPS5J2WJIGO/graph.json","fetch_events":"https://pith.science/api/pith-number/HOQJBI2H4BX76RFZPS5J2WJIGO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO/action/storage_attestation","attest_author":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO/action/author_attestation","sign_citation":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO/action/citation_signature","submit_replication":"https://pith.science/pith/HOQJBI2H4BX76RFZPS5J2WJIGO/action/replication_record"}},"created_at":"2026-05-18T02:50:01.400158+00:00","updated_at":"2026-05-18T02:50:01.400158+00:00"}