{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:JRXW2MU6JGLUN735I4IDH7MTBX","short_pith_number":"pith:JRXW2MU6","schema_version":"1.0","canonical_sha256":"4c6f6d329e499746ff7d471033fd930ded4d4e0f914e4cccd90d833f75c60fb3","source":{"kind":"arxiv","id":"1708.07827","version":2},"attestation_state":"computed","paper":{"title":"Second-Order Optimization for Non-Convex Machine Learning: An Empirical Study","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.NA","stat.ML"],"primary_cat":"math.OC","authors_text":"Farbod Roosta-Khorasani, Michael W. Mahoney, Peng Xu","submitted_at":"2017-08-25T01:33:15Z","abstract_excerpt":"While first-order optimization methods such as stochastic gradient descent (SGD) are popular in machine learning (ML), they come with well-known deficiencies, including relatively-slow convergence, sensitivity to the settings of hyper-parameters such as learning rate, stagnation at high training errors, and difficulty in escaping flat regions and saddle points. These issues are particularly acute in highly non-convex settings such as those arising in neural networks. Motivated by this, there has been recent interest in second-order methods that aim to alleviate these shortcomings by capturing "},"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":"1708.07827","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2017-08-25T01:33:15Z","cross_cats_sorted":["cs.LG","math.NA","stat.ML"],"title_canon_sha256":"d2085d9a14f334607e3f48c430c6d27ee264f5156dc166b61069bd4b8e054bd8","abstract_canon_sha256":"0fb53ce558decae7389d4480981664585a10455aa53248c6fa981e09b1c2f579"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:23:13.996381Z","signature_b64":"ewc4Zn2/EreIGWdbLeG15rKbCBuBcLb7jOg2Ck4MsT5+GY/Dpyi6Cag2bSHJBwbSmZXxkf6jFHACSye/wbmhAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c6f6d329e499746ff7d471033fd930ded4d4e0f914e4cccd90d833f75c60fb3","last_reissued_at":"2026-05-18T00:23:13.995747Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:23:13.995747Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Second-Order Optimization for Non-Convex Machine Learning: An Empirical Study","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.NA","stat.ML"],"primary_cat":"math.OC","authors_text":"Farbod Roosta-Khorasani, Michael W. Mahoney, Peng Xu","submitted_at":"2017-08-25T01:33:15Z","abstract_excerpt":"While first-order optimization methods such as stochastic gradient descent (SGD) are popular in machine learning (ML), they come with well-known deficiencies, including relatively-slow convergence, sensitivity to the settings of hyper-parameters such as learning rate, stagnation at high training errors, and difficulty in escaping flat regions and saddle points. These issues are particularly acute in highly non-convex settings such as those arising in neural networks. Motivated by this, there has been recent interest in second-order methods that aim to alleviate these shortcomings by capturing "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1708.07827","kind":"arxiv","version":2},"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":"1708.07827","created_at":"2026-05-18T00:23:13.995858+00:00"},{"alias_kind":"arxiv_version","alias_value":"1708.07827v2","created_at":"2026-05-18T00:23:13.995858+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1708.07827","created_at":"2026-05-18T00:23:13.995858+00:00"},{"alias_kind":"pith_short_12","alias_value":"JRXW2MU6JGLU","created_at":"2026-05-18T12:31:24.725408+00:00"},{"alias_kind":"pith_short_16","alias_value":"JRXW2MU6JGLUN735","created_at":"2026-05-18T12:31:24.725408+00:00"},{"alias_kind":"pith_short_8","alias_value":"JRXW2MU6","created_at":"2026-05-18T12:31:24.725408+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.13434","citing_title":"Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX","json":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX.json","graph_json":"https://pith.science/api/pith-number/JRXW2MU6JGLUN735I4IDH7MTBX/graph.json","events_json":"https://pith.science/api/pith-number/JRXW2MU6JGLUN735I4IDH7MTBX/events.json","paper":"https://pith.science/paper/JRXW2MU6"},"agent_actions":{"view_html":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX","download_json":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX.json","view_paper":"https://pith.science/paper/JRXW2MU6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1708.07827&json=true","fetch_graph":"https://pith.science/api/pith-number/JRXW2MU6JGLUN735I4IDH7MTBX/graph.json","fetch_events":"https://pith.science/api/pith-number/JRXW2MU6JGLUN735I4IDH7MTBX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX/action/storage_attestation","attest_author":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX/action/author_attestation","sign_citation":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX/action/citation_signature","submit_replication":"https://pith.science/pith/JRXW2MU6JGLUN735I4IDH7MTBX/action/replication_record"}},"created_at":"2026-05-18T00:23:13.995858+00:00","updated_at":"2026-05-18T00:23:13.995858+00:00"}