{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:5BTJUSJ6OONCXT3VNOMMKQLH5K","short_pith_number":"pith:5BTJUSJ6","schema_version":"1.0","canonical_sha256":"e8669a493e739a2bcf756b98c54167ea808da25290166c0ba39d86a881eadb5d","source":{"kind":"arxiv","id":"2003.06307","version":2},"attestation_state":"computed","paper":{"title":"Communication-Efficient Distributed Deep Learning: A Comprehensive Survey","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.SP"],"primary_cat":"cs.DC","authors_text":"Bo Li, Shaohuai Shi, Wei Wang, Xiaowen Chu, Zhenheng Tang","submitted_at":"2020-03-10T05:42:44Z","abstract_excerpt":"Distributed deep learning (DL) has become prevalent in recent years to reduce training time by leveraging multiple computing devices (e.g., GPUs/TPUs) due to larger models and datasets. However, system scalability is limited by communication becoming the performance bottleneck. Addressing this communication issue has become a prominent research topic. In this paper, we provide a comprehensive survey of the communication-efficient distributed training algorithms, focusing on both system-level and algorithmic-level optimizations. We first propose a taxonomy of data-parallel distributed training "},"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":"2003.06307","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2020-03-10T05:42:44Z","cross_cats_sorted":["cs.LG","eess.SP"],"title_canon_sha256":"5fdc5c2707b1eeb1b19462c97ca91b52cce00661202cdc5a913c601f1f40fab3","abstract_canon_sha256":"b4a88b6915b8fa9e926bb1b2ad62abc6a041f9fe72d3ce5165d1364fa62c853d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:46:41.788186Z","signature_b64":"ZIvtvBATsxT2pV9fBHoBnokZ0JJKmuEcTezgpFo1kdhxRnQq+HfzZOSlaPVbBNMCOL9rqFXijWFvm0nlllHpDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8669a493e739a2bcf756b98c54167ea808da25290166c0ba39d86a881eadb5d","last_reissued_at":"2026-07-05T06:46:41.787711Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:46:41.787711Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Communication-Efficient Distributed Deep Learning: A Comprehensive Survey","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.SP"],"primary_cat":"cs.DC","authors_text":"Bo Li, Shaohuai Shi, Wei Wang, Xiaowen Chu, Zhenheng Tang","submitted_at":"2020-03-10T05:42:44Z","abstract_excerpt":"Distributed deep learning (DL) has become prevalent in recent years to reduce training time by leveraging multiple computing devices (e.g., GPUs/TPUs) due to larger models and datasets. However, system scalability is limited by communication becoming the performance bottleneck. Addressing this communication issue has become a prominent research topic. In this paper, we provide a comprehensive survey of the communication-efficient distributed training algorithms, focusing on both system-level and algorithmic-level optimizations. We first propose a taxonomy of data-parallel distributed training "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.06307","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/2003.06307/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":"2003.06307","created_at":"2026-07-05T06:46:41.787772+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.06307v2","created_at":"2026-07-05T06:46:41.787772+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.06307","created_at":"2026-07-05T06:46:41.787772+00:00"},{"alias_kind":"pith_short_12","alias_value":"5BTJUSJ6OONC","created_at":"2026-07-05T06:46:41.787772+00:00"},{"alias_kind":"pith_short_16","alias_value":"5BTJUSJ6OONCXT3V","created_at":"2026-07-05T06:46:41.787772+00:00"},{"alias_kind":"pith_short_8","alias_value":"5BTJUSJ6","created_at":"2026-07-05T06:46:41.787772+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.13434","citing_title":"Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity","ref_index":72,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14993","citing_title":"Serving Chain-structured Jobs with Large Memory Footprints with Application to Large Foundation Model Serving","ref_index":30,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K","json":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K.json","graph_json":"https://pith.science/api/pith-number/5BTJUSJ6OONCXT3VNOMMKQLH5K/graph.json","events_json":"https://pith.science/api/pith-number/5BTJUSJ6OONCXT3VNOMMKQLH5K/events.json","paper":"https://pith.science/paper/5BTJUSJ6"},"agent_actions":{"view_html":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K","download_json":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K.json","view_paper":"https://pith.science/paper/5BTJUSJ6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.06307&json=true","fetch_graph":"https://pith.science/api/pith-number/5BTJUSJ6OONCXT3VNOMMKQLH5K/graph.json","fetch_events":"https://pith.science/api/pith-number/5BTJUSJ6OONCXT3VNOMMKQLH5K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K/action/storage_attestation","attest_author":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K/action/author_attestation","sign_citation":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K/action/citation_signature","submit_replication":"https://pith.science/pith/5BTJUSJ6OONCXT3VNOMMKQLH5K/action/replication_record"}},"created_at":"2026-07-05T06:46:41.787772+00:00","updated_at":"2026-07-05T06:46:41.787772+00:00"}