{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:R4BEEBYOL5UBIVI3JU6QVLRUB3","short_pith_number":"pith:R4BEEBYO","schema_version":"1.0","canonical_sha256":"8f0242070e5f6814551b4d3d0aae340ecca1b50495c76f99d66b6e05e33471f8","source":{"kind":"arxiv","id":"2103.01447","version":3},"attestation_state":"computed","paper":{"title":"ZeroSARAH: Efficient Nonconvex Finite-Sum Optimization with Zero Full Gradient Computation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","math.OC"],"primary_cat":"cs.LG","authors_text":"Peter Richt\\'arik, Slavom\\'ir Hanzely, Zhize Li","submitted_at":"2021-03-02T03:31:06Z","abstract_excerpt":"We propose ZeroSARAH -- a novel variant of the variance-reduced method SARAH (Nguyen et al., 2017) -- for minimizing the average of a large number of nonconvex functions $\\frac{1}{n}\\sum_{i=1}^{n}f_i(x)$. To the best of our knowledge, in this nonconvex finite-sum regime, all existing variance-reduced methods, including SARAH, SVRG, SAGA and their variants, need to compute the full gradient over all $n$ data samples at the initial point $x^0$, and then periodically compute the full gradient once every few iterations (for SVRG, SARAH and their variants). Note that SVRG, SAGA and their variants t"},"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":"2103.01447","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-03-02T03:31:06Z","cross_cats_sorted":["cs.DC","math.OC"],"title_canon_sha256":"a0c58bb4c40efbe9db47713f491c2c0f6e8e88854741a43bf5bbbaa8eb7b9bf6","abstract_canon_sha256":"7bbf61571c32cf9d2391ceb9d16a70418b379500be5ae707b78a7918281a4010"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:21:08.823885Z","signature_b64":"rpLP54MC0aK00sp/ONm1d1whbuwb33Df4YXEtZquZbtYLcwFiREdrPu18Td1nKOwDtCWF60M1ZU14kdsp6ZpAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f0242070e5f6814551b4d3d0aae340ecca1b50495c76f99d66b6e05e33471f8","last_reissued_at":"2026-07-05T03:21:08.823405Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:21:08.823405Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ZeroSARAH: Efficient Nonconvex Finite-Sum Optimization with Zero Full Gradient Computation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","math.OC"],"primary_cat":"cs.LG","authors_text":"Peter Richt\\'arik, Slavom\\'ir Hanzely, Zhize Li","submitted_at":"2021-03-02T03:31:06Z","abstract_excerpt":"We propose ZeroSARAH -- a novel variant of the variance-reduced method SARAH (Nguyen et al., 2017) -- for minimizing the average of a large number of nonconvex functions $\\frac{1}{n}\\sum_{i=1}^{n}f_i(x)$. To the best of our knowledge, in this nonconvex finite-sum regime, all existing variance-reduced methods, including SARAH, SVRG, SAGA and their variants, need to compute the full gradient over all $n$ data samples at the initial point $x^0$, and then periodically compute the full gradient once every few iterations (for SVRG, SARAH and their variants). Note that SVRG, SAGA and their variants t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.01447","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/2103.01447/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":"2103.01447","created_at":"2026-07-05T03:21:08.823461+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.01447v3","created_at":"2026-07-05T03:21:08.823461+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.01447","created_at":"2026-07-05T03:21:08.823461+00:00"},{"alias_kind":"pith_short_12","alias_value":"R4BEEBYOL5UB","created_at":"2026-07-05T03:21:08.823461+00:00"},{"alias_kind":"pith_short_16","alias_value":"R4BEEBYOL5UBIVI3","created_at":"2026-07-05T03:21:08.823461+00:00"},{"alias_kind":"pith_short_8","alias_value":"R4BEEBYO","created_at":"2026-07-05T03:21:08.823461+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.15832","citing_title":"SILAGE: Memory-Efficient, Full-Gradient-Free Nonconvex Optimization for Nested Finite Sums","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20866","citing_title":"LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging","ref_index":169,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18174","citing_title":"Ringmaster LMO: Asynchronous Linear Minimization Oracle Momentum Method","ref_index":167,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13434","citing_title":"Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity","ref_index":272,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08871","citing_title":"Rennala MVR: Improved Time Complexity for Parallel Stochastic Optimization via Momentum-Based Variance Reduction","ref_index":165,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07795","citing_title":"Scalable Distributed Stochastic Optimization via Bidirectional Compression: Beyond Pessimistic Limits","ref_index":85,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3","json":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3.json","graph_json":"https://pith.science/api/pith-number/R4BEEBYOL5UBIVI3JU6QVLRUB3/graph.json","events_json":"https://pith.science/api/pith-number/R4BEEBYOL5UBIVI3JU6QVLRUB3/events.json","paper":"https://pith.science/paper/R4BEEBYO"},"agent_actions":{"view_html":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3","download_json":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3.json","view_paper":"https://pith.science/paper/R4BEEBYO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.01447&json=true","fetch_graph":"https://pith.science/api/pith-number/R4BEEBYOL5UBIVI3JU6QVLRUB3/graph.json","fetch_events":"https://pith.science/api/pith-number/R4BEEBYOL5UBIVI3JU6QVLRUB3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3/action/storage_attestation","attest_author":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3/action/author_attestation","sign_citation":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3/action/citation_signature","submit_replication":"https://pith.science/pith/R4BEEBYOL5UBIVI3JU6QVLRUB3/action/replication_record"}},"created_at":"2026-07-05T03:21:08.823461+00:00","updated_at":"2026-07-05T03:21:08.823461+00:00"}