{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:46NGKA5W6XHVS7U7Z2S7ZHE5PY","short_pith_number":"pith:46NGKA5W","schema_version":"1.0","canonical_sha256":"e79a6503b6f5cf597e9fcea5fc9c9d7e34096f423804e840c3b9e14a99af07ed","source":{"kind":"arxiv","id":"2003.10579","version":1},"attestation_state":"computed","paper":{"title":"Slow and Stale Gradients Can Win the Race","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.LG","cs.PF"],"primary_cat":"stat.ML","authors_text":"Gauri Joshi, Jianyu Wang, Sanghamitra Dutta","submitted_at":"2020-03-23T23:27:50Z","abstract_excerpt":"Distributed Stochastic Gradient Descent (SGD) when run in a synchronous manner, suffers from delays in runtime as it waits for the slowest workers (stragglers). Asynchronous methods can alleviate stragglers, but cause gradient staleness that can adversely affect the convergence error. In this work, we present a novel theoretical characterization of the speedup offered by asynchronous methods by analyzing the trade-off between the error in the trained model and the actual training runtime(wallclock time). The main novelty in our work is that our runtime analysis considers random straggling dela"},"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":"2003.10579","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-03-23T23:27:50Z","cross_cats_sorted":["cs.DC","cs.LG","cs.PF"],"title_canon_sha256":"d52ff666da7fa9863d354b53ff3eb4584187a0d1f12791a456d04fecabd73bf5","abstract_canon_sha256":"865a89d45987213b68e943f04cf9efabbef16343a40b1f03dcdd38d5c450bb14"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:50:18.667896Z","signature_b64":"ktlasyrmI2xJtwvqNRjJ1ecfOuGNx24oP8jiz8jGJWqFfnCedzKyppLXL59iTicHcL1nyYluAkAlV89xjKuLBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e79a6503b6f5cf597e9fcea5fc9c9d7e34096f423804e840c3b9e14a99af07ed","last_reissued_at":"2026-07-05T00:50:18.667356Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:50:18.667356Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Slow and Stale Gradients Can Win the Race","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.LG","cs.PF"],"primary_cat":"stat.ML","authors_text":"Gauri Joshi, Jianyu Wang, Sanghamitra Dutta","submitted_at":"2020-03-23T23:27:50Z","abstract_excerpt":"Distributed Stochastic Gradient Descent (SGD) when run in a synchronous manner, suffers from delays in runtime as it waits for the slowest workers (stragglers). Asynchronous methods can alleviate stragglers, but cause gradient staleness that can adversely affect the convergence error. In this work, we present a novel theoretical characterization of the speedup offered by asynchronous methods by analyzing the trade-off between the error in the trained model and the actual training runtime(wallclock time). The main novelty in our work is that our runtime analysis considers random straggling dela"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.10579","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/2003.10579/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":"2003.10579","created_at":"2026-07-05T00:50:18.667416+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.10579v1","created_at":"2026-07-05T00:50:18.667416+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.10579","created_at":"2026-07-05T00:50:18.667416+00:00"},{"alias_kind":"pith_short_12","alias_value":"46NGKA5W6XHV","created_at":"2026-07-05T00:50:18.667416+00:00"},{"alias_kind":"pith_short_16","alias_value":"46NGKA5W6XHVS7U7","created_at":"2026-07-05T00:50:18.667416+00:00"},{"alias_kind":"pith_short_8","alias_value":"46NGKA5W","created_at":"2026-07-05T00:50:18.667416+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/46NGKA5W6XHVS7U7Z2S7ZHE5PY","json":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY.json","graph_json":"https://pith.science/api/pith-number/46NGKA5W6XHVS7U7Z2S7ZHE5PY/graph.json","events_json":"https://pith.science/api/pith-number/46NGKA5W6XHVS7U7Z2S7ZHE5PY/events.json","paper":"https://pith.science/paper/46NGKA5W"},"agent_actions":{"view_html":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY","download_json":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY.json","view_paper":"https://pith.science/paper/46NGKA5W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.10579&json=true","fetch_graph":"https://pith.science/api/pith-number/46NGKA5W6XHVS7U7Z2S7ZHE5PY/graph.json","fetch_events":"https://pith.science/api/pith-number/46NGKA5W6XHVS7U7Z2S7ZHE5PY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY/action/storage_attestation","attest_author":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY/action/author_attestation","sign_citation":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY/action/citation_signature","submit_replication":"https://pith.science/pith/46NGKA5W6XHVS7U7Z2S7ZHE5PY/action/replication_record"}},"created_at":"2026-07-05T00:50:18.667416+00:00","updated_at":"2026-07-05T00:50:18.667416+00:00"}