{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:XRF4GDLUEJK66HE6HNPICYBRAV","short_pith_number":"pith:XRF4GDLU","canonical_record":{"source":{"id":"1911.08250","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2019-11-19T13:22:22Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"2ff191669b4d66c3422f27553e836e23410901ed44c2217399ed87faa69803f0","abstract_canon_sha256":"9832989c83224ab8dc763d2a219709dd3599840758d587c3066954892cabc91d"},"schema_version":"1.0"},"canonical_sha256":"bc4bc30d742255ef1c9e3b5e816031057533e4dc6a66a08e06008b6864615fc7","source":{"kind":"arxiv","id":"1911.08250","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1911.08250","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"arxiv_version","alias_value":"1911.08250v1","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.08250","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"pith_short_12","alias_value":"XRF4GDLUEJK6","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"pith_short_16","alias_value":"XRF4GDLUEJK66HE6","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"pith_short_8","alias_value":"XRF4GDLU","created_at":"2026-07-05T00:20:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:XRF4GDLUEJK66HE6HNPICYBRAV","target":"record","payload":{"canonical_record":{"source":{"id":"1911.08250","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2019-11-19T13:22:22Z","cross_cats_sorted":["cs.LG","math.OC"],"title_canon_sha256":"2ff191669b4d66c3422f27553e836e23410901ed44c2217399ed87faa69803f0","abstract_canon_sha256":"9832989c83224ab8dc763d2a219709dd3599840758d587c3066954892cabc91d"},"schema_version":"1.0"},"canonical_sha256":"bc4bc30d742255ef1c9e3b5e816031057533e4dc6a66a08e06008b6864615fc7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:20:26.730007Z","signature_b64":"Jx8ykvKvnRNMkXUFwzvamOh90HAvNCTu9MJhECU9p0QBo1nnQmx/mKLbzOo28syprcpha8NODbfKQhOEBFQhDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc4bc30d742255ef1c9e3b5e816031057533e4dc6a66a08e06008b6864615fc7","last_reissued_at":"2026-07-05T00:20:26.729586Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:20:26.729586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1911.08250","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:20:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HOyMQ3S4Twp8ZPmQQhoar939ZLEh20RKsf1KdOR3zcn62NgjGkM82GdPX/iCJVbXUaH6L/czbwjzeqSz7JjgAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:03:46.052250Z"},"content_sha256":"50e1f028e6a5727aa546224b85bb0707c4aab5065c6207bab5800fe71e55555e","schema_version":"1.0","event_id":"sha256:50e1f028e6a5727aa546224b85bb0707c4aab5065c6207bab5800fe71e55555e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:XRF4GDLUEJK66HE6HNPICYBRAV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Discrepancy between the Theoretical Analysis and Practical Implementations of Compressed Communication for Distributed Deep Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","math.OC"],"primary_cat":"cs.DC","authors_text":"Ahmed M. Abdelmoniem, Aritra Dutta, Atal Narayan Sahu, Chen-Yu Ho, El houcine Bergou, Marco Canini, Panos Kalnis","submitted_at":"2019-11-19T13:22:22Z","abstract_excerpt":"Compressed communication, in the form of sparsification or quantization of stochastic gradients, is employed to reduce communication costs in distributed data-parallel training of deep neural networks. However, there exists a discrepancy between theory and practice: while theoretical analysis of most existing compression methods assumes compression is applied to the gradients of the entire model, many practical implementations operate individually on the gradients of each layer of the model. In this paper, we prove that layer-wise compression is, in theory, better, because the convergence rate"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.08250","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/1911.08250/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T00:20:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fPs6bG7kKiwq7Vp850lZkgqg3OXQkjQoOW/xmxiZo2kNVyiwGRV6wB+UjJ9wl1fxaVlDT+9Z69sXC5baSS/SBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:03:46.052989Z"},"content_sha256":"108f1cf9d87c5447c6a2425bd64f42e585af5b77d270a909bf0ddf75eebf9243","schema_version":"1.0","event_id":"sha256:108f1cf9d87c5447c6a2425bd64f42e585af5b77d270a909bf0ddf75eebf9243"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XRF4GDLUEJK66HE6HNPICYBRAV/bundle.json","state_url":"https://pith.science/pith/XRF4GDLUEJK66HE6HNPICYBRAV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XRF4GDLUEJK66HE6HNPICYBRAV/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-07T05:03:46Z","links":{"resolver":"https://pith.science/pith/XRF4GDLUEJK66HE6HNPICYBRAV","bundle":"https://pith.science/pith/XRF4GDLUEJK66HE6HNPICYBRAV/bundle.json","state":"https://pith.science/pith/XRF4GDLUEJK66HE6HNPICYBRAV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XRF4GDLUEJK66HE6HNPICYBRAV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:XRF4GDLUEJK66HE6HNPICYBRAV","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9832989c83224ab8dc763d2a219709dd3599840758d587c3066954892cabc91d","cross_cats_sorted":["cs.LG","math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2019-11-19T13:22:22Z","title_canon_sha256":"2ff191669b4d66c3422f27553e836e23410901ed44c2217399ed87faa69803f0"},"schema_version":"1.0","source":{"id":"1911.08250","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1911.08250","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"arxiv_version","alias_value":"1911.08250v1","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1911.08250","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"pith_short_12","alias_value":"XRF4GDLUEJK6","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"pith_short_16","alias_value":"XRF4GDLUEJK66HE6","created_at":"2026-07-05T00:20:26Z"},{"alias_kind":"pith_short_8","alias_value":"XRF4GDLU","created_at":"2026-07-05T00:20:26Z"}],"graph_snapshots":[{"event_id":"sha256:108f1cf9d87c5447c6a2425bd64f42e585af5b77d270a909bf0ddf75eebf9243","target":"graph","created_at":"2026-07-05T00:20:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/1911.08250/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Compressed communication, in the form of sparsification or quantization of stochastic gradients, is employed to reduce communication costs in distributed data-parallel training of deep neural networks. However, there exists a discrepancy between theory and practice: while theoretical analysis of most existing compression methods assumes compression is applied to the gradients of the entire model, many practical implementations operate individually on the gradients of each layer of the model. In this paper, we prove that layer-wise compression is, in theory, better, because the convergence rate","authors_text":"Ahmed M. Abdelmoniem, Aritra Dutta, Atal Narayan Sahu, Chen-Yu Ho, El houcine Bergou, Marco Canini, Panos Kalnis","cross_cats":["cs.LG","math.OC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2019-11-19T13:22:22Z","title":"On the Discrepancy between the Theoretical Analysis and Practical Implementations of Compressed Communication for Distributed Deep Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1911.08250","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:50e1f028e6a5727aa546224b85bb0707c4aab5065c6207bab5800fe71e55555e","target":"record","created_at":"2026-07-05T00:20:26Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"9832989c83224ab8dc763d2a219709dd3599840758d587c3066954892cabc91d","cross_cats_sorted":["cs.LG","math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2019-11-19T13:22:22Z","title_canon_sha256":"2ff191669b4d66c3422f27553e836e23410901ed44c2217399ed87faa69803f0"},"schema_version":"1.0","source":{"id":"1911.08250","kind":"arxiv","version":1}},"canonical_sha256":"bc4bc30d742255ef1c9e3b5e816031057533e4dc6a66a08e06008b6864615fc7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bc4bc30d742255ef1c9e3b5e816031057533e4dc6a66a08e06008b6864615fc7","first_computed_at":"2026-07-05T00:20:26.729586Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:20:26.729586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Jx8ykvKvnRNMkXUFwzvamOh90HAvNCTu9MJhECU9p0QBo1nnQmx/mKLbzOo28syprcpha8NODbfKQhOEBFQhDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T00:20:26.730007Z","signed_message":"canonical_sha256_bytes"},"source_id":"1911.08250","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:50e1f028e6a5727aa546224b85bb0707c4aab5065c6207bab5800fe71e55555e","sha256:108f1cf9d87c5447c6a2425bd64f42e585af5b77d270a909bf0ddf75eebf9243"],"state_sha256":"54c1f00b192ce35aa86228c72d2df8e6d99445f7185d6b4cc80b9aef57162470"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EyOGBK0buxRpIxPy497Um7mFL24oKbaHsLJBsDNCIiOo/n4GI9i79/ibmTkEk9uT6ygViBhNA0b79Ex8DDCcCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T05:03:46.056086Z","bundle_sha256":"7cdc2a6db4748eb4942b98a1d961c3c617a203db42f259633a75a4fab3d69bc7"}}