{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SACHEHSTH5Y2NFZN77GZK5D4MO","short_pith_number":"pith:SACHEHST","schema_version":"1.0","canonical_sha256":"9004721e533f71a6972dffcd95747c63bc1df333ceb28b394272fee5a2857dd9","source":{"kind":"arxiv","id":"2403.12511","version":3},"attestation_state":"computed","paper":{"title":"Projected Forward Gradient-Guided Frank-Wolfe Algorithm via Variance Reduction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"M. Rostami, S. S. Kia","submitted_at":"2024-03-19T07:25:36Z","abstract_excerpt":"This paper aims to enhance the use of the Frank-Wolfe (FW) algorithm for training deep neural networks. Similar to any gradient-based optimization algorithm, FW suffers from high computational and memory costs when computing gradients for DNNs. This paper introduces the application of the recently proposed projected forward gradient (Projected-FG) method to the FW framework, offering reduced computational cost similar to backpropagation and low memory utilization akin to forward propagation. Our results show that trivial application of the Projected-FG introduces non-vanishing convergence erro"},"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":"2403.12511","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-19T07:25:36Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"ac52263044739dd8d556ea7caaffa283b7dbc423df1e5eec312ee3ee449195fa","abstract_canon_sha256":"ab2c2236dbe258e9eea61d63a1dbd3122f9a6f9d166dd60d563075e252e3ad51"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:54:01.313376Z","signature_b64":"JzMqWVV3lDMN78f96vNSl088wNK0F8bxLa4pRFW9hSgwydwWwEWqOOETmxl6DP3yCw4bpSxBOvi0T/hnhTRxBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9004721e533f71a6972dffcd95747c63bc1df333ceb28b394272fee5a2857dd9","last_reissued_at":"2026-07-05T09:54:01.312916Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:54:01.312916Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Projected Forward Gradient-Guided Frank-Wolfe Algorithm via Variance Reduction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"M. Rostami, S. S. Kia","submitted_at":"2024-03-19T07:25:36Z","abstract_excerpt":"This paper aims to enhance the use of the Frank-Wolfe (FW) algorithm for training deep neural networks. Similar to any gradient-based optimization algorithm, FW suffers from high computational and memory costs when computing gradients for DNNs. This paper introduces the application of the recently proposed projected forward gradient (Projected-FG) method to the FW framework, offering reduced computational cost similar to backpropagation and low memory utilization akin to forward propagation. Our results show that trivial application of the Projected-FG introduces non-vanishing convergence erro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12511","kind":"arxiv","version":3},"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/2403.12511/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":"2403.12511","created_at":"2026-07-05T09:54:01.312972+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.12511v3","created_at":"2026-07-05T09:54:01.312972+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12511","created_at":"2026-07-05T09:54:01.312972+00:00"},{"alias_kind":"pith_short_12","alias_value":"SACHEHSTH5Y2","created_at":"2026-07-05T09:54:01.312972+00:00"},{"alias_kind":"pith_short_16","alias_value":"SACHEHSTH5Y2NFZN","created_at":"2026-07-05T09:54:01.312972+00:00"},{"alias_kind":"pith_short_8","alias_value":"SACHEHST","created_at":"2026-07-05T09:54:01.312972+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2410.02260","citing_title":"FedScalar: Federated Learning with Scalar Communication for Bandwidth-Constrained Networks","ref_index":23,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO","json":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO.json","graph_json":"https://pith.science/api/pith-number/SACHEHSTH5Y2NFZN77GZK5D4MO/graph.json","events_json":"https://pith.science/api/pith-number/SACHEHSTH5Y2NFZN77GZK5D4MO/events.json","paper":"https://pith.science/paper/SACHEHST"},"agent_actions":{"view_html":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO","download_json":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO.json","view_paper":"https://pith.science/paper/SACHEHST","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.12511&json=true","fetch_graph":"https://pith.science/api/pith-number/SACHEHSTH5Y2NFZN77GZK5D4MO/graph.json","fetch_events":"https://pith.science/api/pith-number/SACHEHSTH5Y2NFZN77GZK5D4MO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO/action/storage_attestation","attest_author":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO/action/author_attestation","sign_citation":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO/action/citation_signature","submit_replication":"https://pith.science/pith/SACHEHSTH5Y2NFZN77GZK5D4MO/action/replication_record"}},"created_at":"2026-07-05T09:54:01.312972+00:00","updated_at":"2026-07-05T09:54:01.312972+00:00"}