{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:T4XGT3RV7P5JXRN2GYZEECNJHI","short_pith_number":"pith:T4XGT3RV","schema_version":"1.0","canonical_sha256":"9f2e69ee35fbfa9bc5ba36324209a93a36a5ce8ccf2ba806e525a551a250cadd","source":{"kind":"arxiv","id":"2501.07526","version":2},"attestation_state":"computed","paper":{"title":"Communication-Efficient, 2D Parallel Stochastic Gradient Descent for Distributed-Memory Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.DC","authors_text":"Aditya Devarakonda, Ramakrishnan Kannan","submitted_at":"2025-01-13T17:56:39Z","abstract_excerpt":"Distributed-memory implementations of numerical optimization algorithm, such as stochastic gradient descent (SGD), require interprocessor communication at every iteration of the algorithm. On modern distributed-memory clusters where communication is more expensive than computation, the scalability and performance of these algorithms are limited by communication cost. This work generalizes prior work on 1D $s$-step SGD and 1D Federated SGD with Averaging (FedAvg) to yield a 2D parallel SGD method (HybridSGD) which attains a continuous performance trade off between the two baseline algorithms. W"},"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":"2501.07526","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-01-13T17:56:39Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"565686d34581bb344d4e753cc92f6efbdd67244b4c00f5606b4f257a0fe48abf","abstract_canon_sha256":"4b72b56b8969c98a0843800cf9eaaf25c7cd2818c40ed576e951478f76c40fb9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T01:15:43.213015Z","signature_b64":"/D42AXm41W/UmmIxeVkUw5AcDW56y1lUHGAXk5zA4TXJBiBJvKJ+ouZRfwGHN4stARbjqEJWrpKgo8zJa1KeAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f2e69ee35fbfa9bc5ba36324209a93a36a5ce8ccf2ba806e525a551a250cadd","last_reissued_at":"2026-06-26T01:15:43.212582Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T01:15:43.212582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Communication-Efficient, 2D Parallel Stochastic Gradient Descent for Distributed-Memory Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.DC","authors_text":"Aditya Devarakonda, Ramakrishnan Kannan","submitted_at":"2025-01-13T17:56:39Z","abstract_excerpt":"Distributed-memory implementations of numerical optimization algorithm, such as stochastic gradient descent (SGD), require interprocessor communication at every iteration of the algorithm. On modern distributed-memory clusters where communication is more expensive than computation, the scalability and performance of these algorithms are limited by communication cost. This work generalizes prior work on 1D $s$-step SGD and 1D Federated SGD with Averaging (FedAvg) to yield a 2D parallel SGD method (HybridSGD) which attains a continuous performance trade off between the two baseline algorithms. W"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.07526","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/2501.07526/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":"2501.07526","created_at":"2026-06-26T01:15:43.212641+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.07526v2","created_at":"2026-06-26T01:15:43.212641+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.07526","created_at":"2026-06-26T01:15:43.212641+00:00"},{"alias_kind":"pith_short_12","alias_value":"T4XGT3RV7P5J","created_at":"2026-06-26T01:15:43.212641+00:00"},{"alias_kind":"pith_short_16","alias_value":"T4XGT3RV7P5JXRN2","created_at":"2026-06-26T01:15:43.212641+00:00"},{"alias_kind":"pith_short_8","alias_value":"T4XGT3RV","created_at":"2026-06-26T01:15:43.212641+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/T4XGT3RV7P5JXRN2GYZEECNJHI","json":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI.json","graph_json":"https://pith.science/api/pith-number/T4XGT3RV7P5JXRN2GYZEECNJHI/graph.json","events_json":"https://pith.science/api/pith-number/T4XGT3RV7P5JXRN2GYZEECNJHI/events.json","paper":"https://pith.science/paper/T4XGT3RV"},"agent_actions":{"view_html":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI","download_json":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI.json","view_paper":"https://pith.science/paper/T4XGT3RV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.07526&json=true","fetch_graph":"https://pith.science/api/pith-number/T4XGT3RV7P5JXRN2GYZEECNJHI/graph.json","fetch_events":"https://pith.science/api/pith-number/T4XGT3RV7P5JXRN2GYZEECNJHI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI/action/storage_attestation","attest_author":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI/action/author_attestation","sign_citation":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI/action/citation_signature","submit_replication":"https://pith.science/pith/T4XGT3RV7P5JXRN2GYZEECNJHI/action/replication_record"}},"created_at":"2026-06-26T01:15:43.212641+00:00","updated_at":"2026-06-26T01:15:43.212641+00:00"}