{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:F5MEEHYL7AECMNWZZNP6PTUANT","short_pith_number":"pith:F5MEEHYL","schema_version":"1.0","canonical_sha256":"2f58421f0bf8082636d9cb5fe7ce806cc323e12f4431ce481526f9c8818e6dd1","source":{"kind":"arxiv","id":"2307.11617","version":2},"attestation_state":"computed","paper":{"title":"Robust Fully-Asynchronous Methods for Distributed Training over General Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Jinming Xu, Shibo He, Yan Huang, Ye Tian, Zehan Zhu","submitted_at":"2023-07-21T14:36:40Z","abstract_excerpt":"Perfect synchronization in distributed machine learning problems is inefficient and even impossible due to the existence of latency, package losses and stragglers. We propose a Robust Fully-Asynchronous Stochastic Gradient Tracking method (R-FAST), where each device performs local computation and communication at its own pace without any form of synchronization. Different from existing asynchronous distributed algorithms, R-FAST can eliminate the impact of data heterogeneity across devices and allow for packet losses by employing a robust gradient tracking strategy that relies on properly desi"},"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":"2307.11617","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2023-07-21T14:36:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"99b28b06a6eed4be172440d0f15e48141fb7f06fa35f836e508c21f4dc1b4987","abstract_canon_sha256":"132bf55e09a8b80653c9545b0d5080b58d6f4da65a5d1318fd7201e00b46cc53"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:49:25.096887Z","signature_b64":"hEEg+/Rjd5+2lAzVLmt1FZgQoS4J9cEwrcgqcbb18IHmGFokcxkW1dPkX19tA2KhDf63GkeeC1Y+9xEU1PKXCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f58421f0bf8082636d9cb5fe7ce806cc323e12f4431ce481526f9c8818e6dd1","last_reissued_at":"2026-07-05T08:49:25.096499Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:49:25.096499Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Robust Fully-Asynchronous Methods for Distributed Training over General Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Jinming Xu, Shibo He, Yan Huang, Ye Tian, Zehan Zhu","submitted_at":"2023-07-21T14:36:40Z","abstract_excerpt":"Perfect synchronization in distributed machine learning problems is inefficient and even impossible due to the existence of latency, package losses and stragglers. We propose a Robust Fully-Asynchronous Stochastic Gradient Tracking method (R-FAST), where each device performs local computation and communication at its own pace without any form of synchronization. Different from existing asynchronous distributed algorithms, R-FAST can eliminate the impact of data heterogeneity across devices and allow for packet losses by employing a robust gradient tracking strategy that relies on properly desi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2307.11617","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/2307.11617/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":"2307.11617","created_at":"2026-07-05T08:49:25.096555+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.11617v2","created_at":"2026-07-05T08:49:25.096555+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.11617","created_at":"2026-07-05T08:49:25.096555+00:00"},{"alias_kind":"pith_short_12","alias_value":"F5MEEHYL7AEC","created_at":"2026-07-05T08:49:25.096555+00:00"},{"alias_kind":"pith_short_16","alias_value":"F5MEEHYL7AECMNWZ","created_at":"2026-07-05T08:49:25.096555+00:00"},{"alias_kind":"pith_short_8","alias_value":"F5MEEHYL","created_at":"2026-07-05T08:49:25.096555+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.10322","citing_title":"Asynchronous Decentralized SGD under Non-Convexity: A Block-Coordinate Descent Framework","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT","json":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT.json","graph_json":"https://pith.science/api/pith-number/F5MEEHYL7AECMNWZZNP6PTUANT/graph.json","events_json":"https://pith.science/api/pith-number/F5MEEHYL7AECMNWZZNP6PTUANT/events.json","paper":"https://pith.science/paper/F5MEEHYL"},"agent_actions":{"view_html":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT","download_json":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT.json","view_paper":"https://pith.science/paper/F5MEEHYL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.11617&json=true","fetch_graph":"https://pith.science/api/pith-number/F5MEEHYL7AECMNWZZNP6PTUANT/graph.json","fetch_events":"https://pith.science/api/pith-number/F5MEEHYL7AECMNWZZNP6PTUANT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT/action/storage_attestation","attest_author":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT/action/author_attestation","sign_citation":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT/action/citation_signature","submit_replication":"https://pith.science/pith/F5MEEHYL7AECMNWZZNP6PTUANT/action/replication_record"}},"created_at":"2026-07-05T08:49:25.096555+00:00","updated_at":"2026-07-05T08:49:25.096555+00:00"}