{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:3WNBCR7GPKWXM2TS3OLJB6ZDQ6","short_pith_number":"pith:3WNBCR7G","schema_version":"1.0","canonical_sha256":"dd9a1147e67aad766a72db9690fb2387b1824258d8aab3ad945a0e2e8f8de1d5","source":{"kind":"arxiv","id":"2110.03300","version":1},"attestation_state":"computed","paper":{"title":"Permutation Compressors for Provably Faster Distributed Nonconvex Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Tyurin, Peter Richt\\'arik, Rafa{\\l} Szlendak","submitted_at":"2021-10-07T09:38:15Z","abstract_excerpt":"We study the MARINA method of Gorbunov et al (2021) -- the current state-of-the-art distributed non-convex optimization method in terms of theoretical communication complexity. Theoretical superiority of this method can be largely attributed to two sources: the use of a carefully engineered biased stochastic gradient estimator, which leads to a reduction in the number of communication rounds, and the reliance on {\\em independent} stochastic communication compression operators, which leads to a reduction in the number of transmitted bits within each communication round. In this paper we i) exte"},"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":"2110.03300","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-10-07T09:38:15Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"e07f7b1190784574f732ec73b2d50bd0ed01629f61afa8d42a046559e0f428c0","abstract_canon_sha256":"84c266f9e26f6e2b5500623ac2f8ba5360c0d97d15b3d70564db5b91277da665"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:20:41.304120Z","signature_b64":"YvmY0lPGcoibQu2nUZpKFYRPl6ISv5eJt/9yOO0AMCdYLpvBVph2SvAL1/gP/OY+zDFnFBt9/Qz+b+zKyKVyBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dd9a1147e67aad766a72db9690fb2387b1824258d8aab3ad945a0e2e8f8de1d5","last_reissued_at":"2026-07-05T03:20:41.303690Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:20:41.303690Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Permutation Compressors for Provably Faster Distributed Nonconvex Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alexander Tyurin, Peter Richt\\'arik, Rafa{\\l} Szlendak","submitted_at":"2021-10-07T09:38:15Z","abstract_excerpt":"We study the MARINA method of Gorbunov et al (2021) -- the current state-of-the-art distributed non-convex optimization method in terms of theoretical communication complexity. Theoretical superiority of this method can be largely attributed to two sources: the use of a carefully engineered biased stochastic gradient estimator, which leads to a reduction in the number of communication rounds, and the reliance on {\\em independent} stochastic communication compression operators, which leads to a reduction in the number of transmitted bits within each communication round. In this paper we i) exte"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2110.03300","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/2110.03300/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":"2110.03300","created_at":"2026-07-05T03:20:41.303747+00:00"},{"alias_kind":"arxiv_version","alias_value":"2110.03300v1","created_at":"2026-07-05T03:20:41.303747+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2110.03300","created_at":"2026-07-05T03:20:41.303747+00:00"},{"alias_kind":"pith_short_12","alias_value":"3WNBCR7GPKWX","created_at":"2026-07-05T03:20:41.303747+00:00"},{"alias_kind":"pith_short_16","alias_value":"3WNBCR7GPKWXM2TS","created_at":"2026-07-05T03:20:41.303747+00:00"},{"alias_kind":"pith_short_8","alias_value":"3WNBCR7G","created_at":"2026-07-05T03:20:41.303747+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.20866","citing_title":"LOSCAR-SGD: Local SGD with Communication-Computation Overlap and Delay-Corrected Sparse Model Averaging","ref_index":159,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18174","citing_title":"Ringmaster LMO: Asynchronous Linear Minimization Oracle Momentum Method","ref_index":157,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13434","citing_title":"Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity","ref_index":262,"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":155,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6","json":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6.json","graph_json":"https://pith.science/api/pith-number/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/graph.json","events_json":"https://pith.science/api/pith-number/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/events.json","paper":"https://pith.science/paper/3WNBCR7G"},"agent_actions":{"view_html":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6","download_json":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6.json","view_paper":"https://pith.science/paper/3WNBCR7G","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2110.03300&json=true","fetch_graph":"https://pith.science/api/pith-number/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/graph.json","fetch_events":"https://pith.science/api/pith-number/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/action/storage_attestation","attest_author":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/action/author_attestation","sign_citation":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/action/citation_signature","submit_replication":"https://pith.science/pith/3WNBCR7GPKWXM2TS3OLJB6ZDQ6/action/replication_record"}},"created_at":"2026-07-05T03:20:41.303747+00:00","updated_at":"2026-07-05T03:20:41.303747+00:00"}