{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:56NN6XCKTHADIVEBAVB7VNWNKG","short_pith_number":"pith:56NN6XCK","schema_version":"1.0","canonical_sha256":"ef9adf5c4a99c03454810543fab6cd5182114a7a37d6c31d6886c490ef58822e","source":{"kind":"arxiv","id":"2009.13586","version":6},"attestation_state":"computed","paper":{"title":"Apollo: An Adaptive Parameter-wise Diagonal Quasi-Newton Method for Nonconvex Stochastic Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Xuezhe Ma","submitted_at":"2020-09-28T19:07:02Z","abstract_excerpt":"In this paper, we introduce Apollo, a quasi-Newton method for nonconvex stochastic optimization, which dynamically incorporates the curvature of the loss function by approximating the Hessian via a diagonal matrix. Importantly, the update and storage of the diagonal approximation of Hessian is as efficient as adaptive first-order optimization methods with linear complexity for both time and memory. To handle nonconvexity, we replace the Hessian with its rectified absolute value, which is guaranteed to be positive-definite. Experiments on three tasks of vision and language show that Apollo achi"},"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":"2009.13586","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-09-28T19:07:02Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"08fc6f0fa7ad8fd531308f9d6770547ddb534aae4c4c0eacdae8667ee728d055","abstract_canon_sha256":"da681404257ecad2ad0e59bcafcdcafce924621e6e33eb7150bf081e9709a78a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:07:19.564123Z","signature_b64":"e+U9X0d3/55s5HklkgiwzdNfIISIQENVBua39Mq3w+f1swiLReFqF7fkTWG3uv7oRhS0UzBR7fI5a/x6IyegCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ef9adf5c4a99c03454810543fab6cd5182114a7a37d6c31d6886c490ef58822e","last_reissued_at":"2026-07-05T03:07:19.563477Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:07:19.563477Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Apollo: An Adaptive Parameter-wise Diagonal Quasi-Newton Method for Nonconvex Stochastic Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Xuezhe Ma","submitted_at":"2020-09-28T19:07:02Z","abstract_excerpt":"In this paper, we introduce Apollo, a quasi-Newton method for nonconvex stochastic optimization, which dynamically incorporates the curvature of the loss function by approximating the Hessian via a diagonal matrix. Importantly, the update and storage of the diagonal approximation of Hessian is as efficient as adaptive first-order optimization methods with linear complexity for both time and memory. To handle nonconvexity, we replace the Hessian with its rectified absolute value, which is guaranteed to be positive-definite. Experiments on three tasks of vision and language show that Apollo achi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.13586","kind":"arxiv","version":6},"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/2009.13586/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":"2009.13586","created_at":"2026-07-05T03:07:19.563546+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.13586v6","created_at":"2026-07-05T03:07:19.563546+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.13586","created_at":"2026-07-05T03:07:19.563546+00:00"},{"alias_kind":"pith_short_12","alias_value":"56NN6XCKTHAD","created_at":"2026-07-05T03:07:19.563546+00:00"},{"alias_kind":"pith_short_16","alias_value":"56NN6XCKTHADIVEB","created_at":"2026-07-05T03:07:19.563546+00:00"},{"alias_kind":"pith_short_8","alias_value":"56NN6XCK","created_at":"2026-07-05T03:07:19.563546+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.08233","citing_title":"Strategies for Improving Communication Efficiency in Distributed and Federated Learning: Compression, Local Training, and Personalization","ref_index":149,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG","json":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG.json","graph_json":"https://pith.science/api/pith-number/56NN6XCKTHADIVEBAVB7VNWNKG/graph.json","events_json":"https://pith.science/api/pith-number/56NN6XCKTHADIVEBAVB7VNWNKG/events.json","paper":"https://pith.science/paper/56NN6XCK"},"agent_actions":{"view_html":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG","download_json":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG.json","view_paper":"https://pith.science/paper/56NN6XCK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.13586&json=true","fetch_graph":"https://pith.science/api/pith-number/56NN6XCKTHADIVEBAVB7VNWNKG/graph.json","fetch_events":"https://pith.science/api/pith-number/56NN6XCKTHADIVEBAVB7VNWNKG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG/action/storage_attestation","attest_author":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG/action/author_attestation","sign_citation":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG/action/citation_signature","submit_replication":"https://pith.science/pith/56NN6XCKTHADIVEBAVB7VNWNKG/action/replication_record"}},"created_at":"2026-07-05T03:07:19.563546+00:00","updated_at":"2026-07-05T03:07:19.563546+00:00"}