{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:7WNKV26EPLVY3KJIM74CJRBWS4","short_pith_number":"pith:7WNKV26E","schema_version":"1.0","canonical_sha256":"fd9aaaebc47aeb8da92867f824c436972cfb522d2bbaaef994e443a9102e4abd","source":{"kind":"arxiv","id":"2012.12418","version":2},"attestation_state":"computed","paper":{"title":"Stochastic Gradient Variance Reduction by Solving a Filtering Problem","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Xingyi Yang","submitted_at":"2020-12-22T23:48:42Z","abstract_excerpt":"Deep neural networks (DNN) are typically optimized using stochastic gradient descent (SGD). However, the estimation of the gradient using stochastic samples tends to be noisy and unreliable, resulting in large gradient variance and bad convergence. In this paper, we propose \\textbf{Filter Gradient Decent}~(FGD), an efficient stochastic optimization algorithm that makes the consistent estimation of the local gradient by solving an adaptive filtering problem with different design of filters. Our method reduces variance in stochastic gradient descent by incorporating the historical states to enha"},"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":"2012.12418","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-12-22T23:48:42Z","cross_cats_sorted":["cs.CV","stat.ML"],"title_canon_sha256":"38ff54f2697eb499a42b581d81550509fcdc3b75db8c90dfff179e322ccf4716","abstract_canon_sha256":"d8982ac02deb6045bf6b4c06a40401fa0e1e9fc5c2041a74066fd3bec5ee50b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:40:30.887268Z","signature_b64":"7eNSCX3cw0evXEQz5YOReBYmk0LdgQ3qpHAGXvHrtTuyHkdsoSNZG3aVwkJGbA+WsvUdIbPeCM9Ps0tkyuqOAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fd9aaaebc47aeb8da92867f824c436972cfb522d2bbaaef994e443a9102e4abd","last_reissued_at":"2026-07-05T02:40:30.886910Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:40:30.886910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stochastic Gradient Variance Reduction by Solving a Filtering Problem","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Xingyi Yang","submitted_at":"2020-12-22T23:48:42Z","abstract_excerpt":"Deep neural networks (DNN) are typically optimized using stochastic gradient descent (SGD). However, the estimation of the gradient using stochastic samples tends to be noisy and unreliable, resulting in large gradient variance and bad convergence. In this paper, we propose \\textbf{Filter Gradient Decent}~(FGD), an efficient stochastic optimization algorithm that makes the consistent estimation of the local gradient by solving an adaptive filtering problem with different design of filters. Our method reduces variance in stochastic gradient descent by incorporating the historical states to enha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2012.12418","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2012.12418/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":"2012.12418","created_at":"2026-07-05T02:40:30.886959+00:00"},{"alias_kind":"arxiv_version","alias_value":"2012.12418v2","created_at":"2026-07-05T02:40:30.886959+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2012.12418","created_at":"2026-07-05T02:40:30.886959+00:00"},{"alias_kind":"pith_short_12","alias_value":"7WNKV26EPLVY","created_at":"2026-07-05T02:40:30.886959+00:00"},{"alias_kind":"pith_short_16","alias_value":"7WNKV26EPLVY3KJI","created_at":"2026-07-05T02:40:30.886959+00:00"},{"alias_kind":"pith_short_8","alias_value":"7WNKV26E","created_at":"2026-07-05T02:40:30.886959+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.15388","citing_title":"Unified High-Probability Analysis of Stochastic Variance-Reduced Estimation","ref_index":143,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4","json":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4.json","graph_json":"https://pith.science/api/pith-number/7WNKV26EPLVY3KJIM74CJRBWS4/graph.json","events_json":"https://pith.science/api/pith-number/7WNKV26EPLVY3KJIM74CJRBWS4/events.json","paper":"https://pith.science/paper/7WNKV26E"},"agent_actions":{"view_html":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4","download_json":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4.json","view_paper":"https://pith.science/paper/7WNKV26E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2012.12418&json=true","fetch_graph":"https://pith.science/api/pith-number/7WNKV26EPLVY3KJIM74CJRBWS4/graph.json","fetch_events":"https://pith.science/api/pith-number/7WNKV26EPLVY3KJIM74CJRBWS4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4/action/storage_attestation","attest_author":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4/action/author_attestation","sign_citation":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4/action/citation_signature","submit_replication":"https://pith.science/pith/7WNKV26EPLVY3KJIM74CJRBWS4/action/replication_record"}},"created_at":"2026-07-05T02:40:30.886959+00:00","updated_at":"2026-07-05T02:40:30.886959+00:00"}