{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:2LD2KJNQBHKB7OK357CYQEXRCP","short_pith_number":"pith:2LD2KJNQ","schema_version":"1.0","canonical_sha256":"d2c7a525b009d41fb95befc58812f113c173c56672de66adef1978e2b57f933b","source":{"kind":"arxiv","id":"2206.15143","version":1},"attestation_state":"computed","paper":{"title":"Scalable K-FAC Training for Deep Neural Networks with Distributed Preconditioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Bo Li, Lin Zhang, Shaohuai Shi, Wei Wang","submitted_at":"2022-06-30T09:22:25Z","abstract_excerpt":"The second-order optimization methods, notably the D-KFAC (Distributed Kronecker Factored Approximate Curvature) algorithms, have gained traction on accelerating deep neural network (DNN) training on GPU clusters. However, existing D-KFAC algorithms require to compute and communicate a large volume of second-order information, i.e., Kronecker factors (KFs), before preconditioning gradients, resulting in large computation and communication overheads as well as a high memory footprint. In this paper, we propose DP-KFAC, a novel distributed preconditioning scheme that distributes the KF construct"},"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":"2206.15143","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-30T09:22:25Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"0854a5c609265f53d6bafadcc326213f1e07331e40968dda2a7f580b1d48c09f","abstract_canon_sha256":"de3340a33196e2c17020e1140ef26b932b595067d2254191d03c6ffbc068d6b5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:36:25.858141Z","signature_b64":"dkv6EgyOYuUYCiaWcvjodqUja8K2ZQTqSLtcVQ2RPcnZAhO9I23MfA4mh80HvqSYQs+8shKf08h9u7T+gvR7Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d2c7a525b009d41fb95befc58812f113c173c56672de66adef1978e2b57f933b","last_reissued_at":"2026-07-05T04:36:25.857751Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:36:25.857751Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable K-FAC Training for Deep Neural Networks with Distributed Preconditioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Bo Li, Lin Zhang, Shaohuai Shi, Wei Wang","submitted_at":"2022-06-30T09:22:25Z","abstract_excerpt":"The second-order optimization methods, notably the D-KFAC (Distributed Kronecker Factored Approximate Curvature) algorithms, have gained traction on accelerating deep neural network (DNN) training on GPU clusters. However, existing D-KFAC algorithms require to compute and communicate a large volume of second-order information, i.e., Kronecker factors (KFs), before preconditioning gradients, resulting in large computation and communication overheads as well as a high memory footprint. In this paper, we propose DP-KFAC, a novel distributed preconditioning scheme that distributes the KF construct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.15143","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2206.15143/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":"2206.15143","created_at":"2026-07-05T04:36:25.857815+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.15143v1","created_at":"2026-07-05T04:36:25.857815+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.15143","created_at":"2026-07-05T04:36:25.857815+00:00"},{"alias_kind":"pith_short_12","alias_value":"2LD2KJNQBHKB","created_at":"2026-07-05T04:36:25.857815+00:00"},{"alias_kind":"pith_short_16","alias_value":"2LD2KJNQBHKB7OK3","created_at":"2026-07-05T04:36:25.857815+00:00"},{"alias_kind":"pith_short_8","alias_value":"2LD2KJNQ","created_at":"2026-07-05T04:36:25.857815+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP","json":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP.json","graph_json":"https://pith.science/api/pith-number/2LD2KJNQBHKB7OK357CYQEXRCP/graph.json","events_json":"https://pith.science/api/pith-number/2LD2KJNQBHKB7OK357CYQEXRCP/events.json","paper":"https://pith.science/paper/2LD2KJNQ"},"agent_actions":{"view_html":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP","download_json":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP.json","view_paper":"https://pith.science/paper/2LD2KJNQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.15143&json=true","fetch_graph":"https://pith.science/api/pith-number/2LD2KJNQBHKB7OK357CYQEXRCP/graph.json","fetch_events":"https://pith.science/api/pith-number/2LD2KJNQBHKB7OK357CYQEXRCP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP/action/storage_attestation","attest_author":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP/action/author_attestation","sign_citation":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP/action/citation_signature","submit_replication":"https://pith.science/pith/2LD2KJNQBHKB7OK357CYQEXRCP/action/replication_record"}},"created_at":"2026-07-05T04:36:25.857815+00:00","updated_at":"2026-07-05T04:36:25.857815+00:00"}