{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:QRPQHNMFV7DEXS7HYR5ILQ5FCZ","short_pith_number":"pith:QRPQHNMF","canonical_record":{"source":{"id":"2002.08056","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-19T08:45:54Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"1f8419284b4b86c830f55d13c99d20daa4adaf237d886c5c916fc8eeb83d8770","abstract_canon_sha256":"d24f1f0339e3a38f173ee3e86e8ac14f8eca34dce3ca3ad4328925239d6f0730"},"schema_version":"1.0"},"canonical_sha256":"845f03b585afc64bcbe7c47a85c3a51678544e91743aead9fbd7d95dd6d1007b","source":{"kind":"arxiv","id":"2002.08056","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2002.08056","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"arxiv_version","alias_value":"2002.08056v1","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08056","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"pith_short_12","alias_value":"QRPQHNMFV7DE","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"pith_short_16","alias_value":"QRPQHNMFV7DEXS7H","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"pith_short_8","alias_value":"QRPQHNMF","created_at":"2026-07-05T00:42:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:QRPQHNMFV7DEXS7HYR5ILQ5FCZ","target":"record","payload":{"canonical_record":{"source":{"id":"2002.08056","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-19T08:45:54Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"1f8419284b4b86c830f55d13c99d20daa4adaf237d886c5c916fc8eeb83d8770","abstract_canon_sha256":"d24f1f0339e3a38f173ee3e86e8ac14f8eca34dce3ca3ad4328925239d6f0730"},"schema_version":"1.0"},"canonical_sha256":"845f03b585afc64bcbe7c47a85c3a51678544e91743aead9fbd7d95dd6d1007b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:42:25.957618Z","signature_b64":"xAyMyZLFhOxdlQD/m7e+4q24C/JEWzGS/mmah0JPKU5pi0hEWh1IK8BmuLeEKlnA8Paps7LJI4AqVpt1Q/eNBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"845f03b585afc64bcbe7c47a85c3a51678544e91743aead9fbd7d95dd6d1007b","last_reissued_at":"2026-07-05T00:42:25.957214Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:42:25.957214Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2002.08056","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":"2026-07-05T00:42:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OrYcdEGPCXicYwsrpodeyTOvhXQ78895d8ucaZbhkOCN6vy/0hvddO5zSqCp/7kX4scKYMnOofVIUDrfZIk4BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T07:43:22.037305Z"},"content_sha256":"9e395914c78a689dbd28bd9f900243678c60f756e7d82e3c579fb9394c943f3c","schema_version":"1.0","event_id":"sha256:9e395914c78a689dbd28bd9f900243678c60f756e7d82e3c579fb9394c943f3c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:QRPQHNMFV7DEXS7HYR5ILQ5FCZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Geometry of Sign Gradient Descent","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Fabian Pedregosa, Lukas Balles, Nicolas Le Roux","submitted_at":"2020-02-19T08:45:54Z","abstract_excerpt":"Sign-based optimization methods have become popular in machine learning due to their favorable communication cost in distributed optimization and their surprisingly good performance in neural network training. Furthermore, they are closely connected to so-called adaptive gradient methods like Adam. Recent works on signSGD have used a non-standard \"separable smoothness\" assumption, whereas some older works study sign gradient descent as steepest descent with respect to the $\\ell_\\infty$-norm. In this work, we unify these existing results by showing a close connection between separable smoothnes"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08056","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/2002.08056/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":"2026-07-05T00:42:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pZz86dmUuiD8UQeo1U0hGbfBMLaShnTvsowaQyFb6ZQGO5Yd8zbopHrchJCyJ+bKslF6BlWBfyKmM3UmkuIrCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T07:43:22.037819Z"},"content_sha256":"25d579e7258f5b90b9ea906bd1d9e7d4f1959fd6fd996b714d22c0fc7a06d086","schema_version":"1.0","event_id":"sha256:25d579e7258f5b90b9ea906bd1d9e7d4f1959fd6fd996b714d22c0fc7a06d086"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QRPQHNMFV7DEXS7HYR5ILQ5FCZ/bundle.json","state_url":"https://pith.science/pith/QRPQHNMFV7DEXS7HYR5ILQ5FCZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QRPQHNMFV7DEXS7HYR5ILQ5FCZ/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":"2026-08-04T07:43:22Z","links":{"resolver":"https://pith.science/pith/QRPQHNMFV7DEXS7HYR5ILQ5FCZ","bundle":"https://pith.science/pith/QRPQHNMFV7DEXS7HYR5ILQ5FCZ/bundle.json","state":"https://pith.science/pith/QRPQHNMFV7DEXS7HYR5ILQ5FCZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QRPQHNMFV7DEXS7HYR5ILQ5FCZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:QRPQHNMFV7DEXS7HYR5ILQ5FCZ","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":"d24f1f0339e3a38f173ee3e86e8ac14f8eca34dce3ca3ad4328925239d6f0730","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-19T08:45:54Z","title_canon_sha256":"1f8419284b4b86c830f55d13c99d20daa4adaf237d886c5c916fc8eeb83d8770"},"schema_version":"1.0","source":{"id":"2002.08056","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2002.08056","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"arxiv_version","alias_value":"2002.08056v1","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2002.08056","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"pith_short_12","alias_value":"QRPQHNMFV7DE","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"pith_short_16","alias_value":"QRPQHNMFV7DEXS7H","created_at":"2026-07-05T00:42:25Z"},{"alias_kind":"pith_short_8","alias_value":"QRPQHNMF","created_at":"2026-07-05T00:42:25Z"}],"graph_snapshots":[{"event_id":"sha256:25d579e7258f5b90b9ea906bd1d9e7d4f1959fd6fd996b714d22c0fc7a06d086","target":"graph","created_at":"2026-07-05T00:42:25Z","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/2002.08056/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Sign-based optimization methods have become popular in machine learning due to their favorable communication cost in distributed optimization and their surprisingly good performance in neural network training. Furthermore, they are closely connected to so-called adaptive gradient methods like Adam. Recent works on signSGD have used a non-standard \"separable smoothness\" assumption, whereas some older works study sign gradient descent as steepest descent with respect to the $\\ell_\\infty$-norm. In this work, we unify these existing results by showing a close connection between separable smoothnes","authors_text":"Fabian Pedregosa, Lukas Balles, Nicolas Le Roux","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-19T08:45:54Z","title":"The Geometry of Sign Gradient Descent"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2002.08056","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":"sha256:9e395914c78a689dbd28bd9f900243678c60f756e7d82e3c579fb9394c943f3c","target":"record","created_at":"2026-07-05T00:42:25Z","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":"d24f1f0339e3a38f173ee3e86e8ac14f8eca34dce3ca3ad4328925239d6f0730","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-02-19T08:45:54Z","title_canon_sha256":"1f8419284b4b86c830f55d13c99d20daa4adaf237d886c5c916fc8eeb83d8770"},"schema_version":"1.0","source":{"id":"2002.08056","kind":"arxiv","version":1}},"canonical_sha256":"845f03b585afc64bcbe7c47a85c3a51678544e91743aead9fbd7d95dd6d1007b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"845f03b585afc64bcbe7c47a85c3a51678544e91743aead9fbd7d95dd6d1007b","first_computed_at":"2026-07-05T00:42:25.957214Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:42:25.957214Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xAyMyZLFhOxdlQD/m7e+4q24C/JEWzGS/mmah0JPKU5pi0hEWh1IK8BmuLeEKlnA8Paps7LJI4AqVpt1Q/eNBA==","signature_status":"signed_v1","signed_at":"2026-07-05T00:42:25.957618Z","signed_message":"canonical_sha256_bytes"},"source_id":"2002.08056","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9e395914c78a689dbd28bd9f900243678c60f756e7d82e3c579fb9394c943f3c","sha256:25d579e7258f5b90b9ea906bd1d9e7d4f1959fd6fd996b714d22c0fc7a06d086"],"state_sha256":"f63f3d9ef78f91a33979c599178b973ca8713f0ef124c3a8dc6da37d2645d8f1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ESaP7/gR5GOFtQWcQgF/M2AzpZCCINspq3f1qInwNmj7yYemeX86+eTQ4OWmmLUT1osBGm/F+HC8HbcvNlkhDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T07:43:22.041198Z","bundle_sha256":"48782941c9d877c30fdd02a0e6d7b3d1fc503edc6f445ed751c62abb7c3684d4"}}