{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:JIQOTZ7JYZAS6XNH7VFAP3RK5P","short_pith_number":"pith:JIQOTZ7J","schema_version":"1.0","canonical_sha256":"4a20e9e7e9c6412f5da7fd4a07ee2aebeb064042fbfb6f07e374a1bc2c95a21f","source":{"kind":"arxiv","id":"1901.09269","version":3},"attestation_state":"computed","paper":{"title":"Distributed Learning with Compressed Gradient Differences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Eduard Gorbunov, Konstantin Mishchenko, Martin Tak\\'a\\v{c}, Peter Richt\\'arik","submitted_at":"2019-01-26T19:50:45Z","abstract_excerpt":"Training large machine learning models requires a distributed computing approach, with communication of the model updates being the bottleneck. For this reason, several methods based on the compression (e.g., sparsification and/or quantization) of updates were recently proposed, including QSGD (Alistarh et al., 2017), TernGrad (Wen et al., 2017), SignSGD (Bernstein et al., 2018), and DQGD (Khirirat et al., 2018). However, none of these methods are able to learn the gradients, which renders them incapable of converging to the true optimum in the batch mode. In this work we propose a new distrib"},"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":"1901.09269","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-01-26T19:50:45Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"3f34e856ce9a9ea20c9f9131c37b55a5b0573136eecaae1f2bf7fc8cd822bae5","abstract_canon_sha256":"8582d49ad493ac035006c07ecb0a9840d081a8beedddf1437bdb87052b7577a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:28:24.389103Z","signature_b64":"it8mvNX7vhmS/O6ddDcTUqAwAWlDkkjaMKBg8yOSs6Ww7RquOXLBu+gqYWa1X+elLFi05xKOcjOrNUqMNDHRCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a20e9e7e9c6412f5da7fd4a07ee2aebeb064042fbfb6f07e374a1bc2c95a21f","last_reissued_at":"2026-07-05T07:28:24.388631Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:28:24.388631Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distributed Learning with Compressed Gradient Differences","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Eduard Gorbunov, Konstantin Mishchenko, Martin Tak\\'a\\v{c}, Peter Richt\\'arik","submitted_at":"2019-01-26T19:50:45Z","abstract_excerpt":"Training large machine learning models requires a distributed computing approach, with communication of the model updates being the bottleneck. For this reason, several methods based on the compression (e.g., sparsification and/or quantization) of updates were recently proposed, including QSGD (Alistarh et al., 2017), TernGrad (Wen et al., 2017), SignSGD (Bernstein et al., 2018), and DQGD (Khirirat et al., 2018). However, none of these methods are able to learn the gradients, which renders them incapable of converging to the true optimum in the batch mode. In this work we propose a new distrib"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1901.09269","kind":"arxiv","version":3},"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/1901.09269/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":"1901.09269","created_at":"2026-07-05T07:28:24.388699+00:00"},{"alias_kind":"arxiv_version","alias_value":"1901.09269v3","created_at":"2026-07-05T07:28:24.388699+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1901.09269","created_at":"2026-07-05T07:28:24.388699+00:00"},{"alias_kind":"pith_short_12","alias_value":"JIQOTZ7JYZAS","created_at":"2026-07-05T07:28:24.388699+00:00"},{"alias_kind":"pith_short_16","alias_value":"JIQOTZ7JYZAS6XNH","created_at":"2026-07-05T07:28:24.388699+00:00"},{"alias_kind":"pith_short_8","alias_value":"JIQOTZ7J","created_at":"2026-07-05T07:28:24.388699+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"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":99,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18174","citing_title":"Ringmaster LMO: Asynchronous Linear Minimization Oracle Momentum Method","ref_index":97,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13434","citing_title":"Rescaled Asynchronous SGD: Optimal Distributed Optimization under Data and System Heterogeneity","ref_index":213,"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":95,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07795","citing_title":"Scalable Distributed Stochastic Optimization via Bidirectional Compression: Beyond Pessimistic Limits","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14663","citing_title":"EdgeDetect: Importance-Aware Gradient Compression with Homomorphic Aggregation for Federated Intrusion Detection","ref_index":43,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P","json":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P.json","graph_json":"https://pith.science/api/pith-number/JIQOTZ7JYZAS6XNH7VFAP3RK5P/graph.json","events_json":"https://pith.science/api/pith-number/JIQOTZ7JYZAS6XNH7VFAP3RK5P/events.json","paper":"https://pith.science/paper/JIQOTZ7J"},"agent_actions":{"view_html":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P","download_json":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P.json","view_paper":"https://pith.science/paper/JIQOTZ7J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1901.09269&json=true","fetch_graph":"https://pith.science/api/pith-number/JIQOTZ7JYZAS6XNH7VFAP3RK5P/graph.json","fetch_events":"https://pith.science/api/pith-number/JIQOTZ7JYZAS6XNH7VFAP3RK5P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P/action/storage_attestation","attest_author":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P/action/author_attestation","sign_citation":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P/action/citation_signature","submit_replication":"https://pith.science/pith/JIQOTZ7JYZAS6XNH7VFAP3RK5P/action/replication_record"}},"created_at":"2026-07-05T07:28:24.388699+00:00","updated_at":"2026-07-05T07:28:24.388699+00:00"}