{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:SZMPYNXEZRGKA4WSTJ6SHLMCLD","short_pith_number":"pith:SZMPYNXE","schema_version":"1.0","canonical_sha256":"9658fc36e4cc4ca072d29a7d23ad8258cbd683dfdb9a98537952c5f4fa401b10","source":{"kind":"arxiv","id":"1809.08694","version":5},"attestation_state":"computed","paper":{"title":"Second-order Guarantees of Distributed Gradient Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"math.OC","authors_text":"Amir Daneshmand, Gesualdo Scutari, Vyacheslav Kungurtsev","submitted_at":"2018-09-23T23:06:39Z","abstract_excerpt":"We consider distributed smooth nonconvex unconstrained optimization over networks, modeled as a connected graph. We examine the behavior of distributed gradient-based algorithms near strict saddle points. Specifically, we establish that (i) the renowned Distributed Gradient Descent (DGD) algorithm likely converges to a neighborhood of a Second-order Stationary (SoS) solution; and (ii) the more recent class of distributed algorithms based on gradient tracking--implementable also over digraphs--likely converges to exact SoS solutions, thus avoiding (strict) saddle-points. Furthermore, new conver"},"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":"1809.08694","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2018-09-23T23:06:39Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"dca32a012b81bc12ae10dbbf3583be4d3cd3bc350d245483d9491206319a62c3","abstract_canon_sha256":"a1a8dd09e79f44c23f4817088070acb3ba00597426c1a04c629fcff8555abf44"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:05:22.788996Z","signature_b64":"TFJfzj7U2qS9VExRzRV27PaeNYksu7g/kg2AGGqTgZVW4YuDXBvowBs7u/VfY1VvaKfG5Iz9fpNoFH6QU/IsAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9658fc36e4cc4ca072d29a7d23ad8258cbd683dfdb9a98537952c5f4fa401b10","last_reissued_at":"2026-07-05T01:05:22.788491Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:05:22.788491Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Second-order Guarantees of Distributed Gradient Algorithms","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"math.OC","authors_text":"Amir Daneshmand, Gesualdo Scutari, Vyacheslav Kungurtsev","submitted_at":"2018-09-23T23:06:39Z","abstract_excerpt":"We consider distributed smooth nonconvex unconstrained optimization over networks, modeled as a connected graph. We examine the behavior of distributed gradient-based algorithms near strict saddle points. Specifically, we establish that (i) the renowned Distributed Gradient Descent (DGD) algorithm likely converges to a neighborhood of a Second-order Stationary (SoS) solution; and (ii) the more recent class of distributed algorithms based on gradient tracking--implementable also over digraphs--likely converges to exact SoS solutions, thus avoiding (strict) saddle-points. Furthermore, new conver"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1809.08694","kind":"arxiv","version":5},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/1809.08694/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"1809.08694","created_at":"2026-07-05T01:05:22.788552+00:00"},{"alias_kind":"arxiv_version","alias_value":"1809.08694v5","created_at":"2026-07-05T01:05:22.788552+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1809.08694","created_at":"2026-07-05T01:05:22.788552+00:00"},{"alias_kind":"pith_short_12","alias_value":"SZMPYNXEZRGK","created_at":"2026-07-05T01:05:22.788552+00:00"},{"alias_kind":"pith_short_16","alias_value":"SZMPYNXEZRGKA4WS","created_at":"2026-07-05T01:05:22.788552+00:00"},{"alias_kind":"pith_short_8","alias_value":"SZMPYNXE","created_at":"2026-07-05T01:05:22.788552+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"1907.01848","citing_title":"Distributed Learning in Non-Convex Environments -- Part I: Agreement at a Linear Rate","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"1907.01849","citing_title":"Distributed Learning in Non-Convex Environments -- Part II: Polynomial Escape from Saddle-Points","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"1907.08802","citing_title":"Distributed Global Optimization by Annealing","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD","json":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD.json","graph_json":"https://pith.science/api/pith-number/SZMPYNXEZRGKA4WSTJ6SHLMCLD/graph.json","events_json":"https://pith.science/api/pith-number/SZMPYNXEZRGKA4WSTJ6SHLMCLD/events.json","paper":"https://pith.science/paper/SZMPYNXE"},"agent_actions":{"view_html":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD","download_json":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD.json","view_paper":"https://pith.science/paper/SZMPYNXE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1809.08694&json=true","fetch_graph":"https://pith.science/api/pith-number/SZMPYNXEZRGKA4WSTJ6SHLMCLD/graph.json","fetch_events":"https://pith.science/api/pith-number/SZMPYNXEZRGKA4WSTJ6SHLMCLD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD/action/storage_attestation","attest_author":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD/action/author_attestation","sign_citation":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD/action/citation_signature","submit_replication":"https://pith.science/pith/SZMPYNXEZRGKA4WSTJ6SHLMCLD/action/replication_record"}},"created_at":"2026-07-05T01:05:22.788552+00:00","updated_at":"2026-07-05T01:05:22.788552+00:00"}