{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:CTADMPSUWHKYMQ3ZPAGTGPGMMV","short_pith_number":"pith:CTADMPSU","canonical_record":{"source":{"id":"2410.13849","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2024-10-17T17:59:01Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"c9f89eedac83b4f31f9921cf092f8cceb57b1c0699e5f7c58860d4d6ebb06248","abstract_canon_sha256":"9915a3a2caf3a444dd149a57793e13d2585cd1aec3bb11ab631a412b997433c2"},"schema_version":"1.0"},"canonical_sha256":"14c0363e54b1d5864379780d333ccc6553b1968a306211af177db9981d6dfe3e","source":{"kind":"arxiv","id":"2410.13849","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.13849","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"arxiv_version","alias_value":"2410.13849v3","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13849","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"pith_short_12","alias_value":"CTADMPSUWHKY","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"pith_short_16","alias_value":"CTADMPSUWHKYMQ3Z","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"pith_short_8","alias_value":"CTADMPSU","created_at":"2026-07-05T11:34:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:CTADMPSUWHKYMQ3ZPAGTGPGMMV","target":"record","payload":{"canonical_record":{"source":{"id":"2410.13849","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2024-10-17T17:59:01Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"c9f89eedac83b4f31f9921cf092f8cceb57b1c0699e5f7c58860d4d6ebb06248","abstract_canon_sha256":"9915a3a2caf3a444dd149a57793e13d2585cd1aec3bb11ab631a412b997433c2"},"schema_version":"1.0"},"canonical_sha256":"14c0363e54b1d5864379780d333ccc6553b1968a306211af177db9981d6dfe3e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:34:00.673420Z","signature_b64":"ogcx6xzaYZ3xz+eZd0wrDipGEp0d1qsQKvyUJEiiNBLEOPuKTH1w1HNdGQA25jkZXgzBdYmXSUu8T7nSSsLZCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"14c0363e54b1d5864379780d333ccc6553b1968a306211af177db9981d6dfe3e","last_reissued_at":"2026-07-05T11:34:00.672793Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:34:00.672793Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.13849","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:34:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6lupub2FvXKsASexkAfZPsqZr5V60JN0RAo9K7mIbX5ke9kyR9egLrM5t318XECUalcH5YmrJ8kammxasEDdCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T20:53:22.570249Z"},"content_sha256":"f139168830d7ea36181155061bb792755c6daefb6a13a87a58fc9ae440466b70","schema_version":"1.0","event_id":"sha256:f139168830d7ea36181155061bb792755c6daefb6a13a87a58fc9ae440466b70"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:CTADMPSUWHKYMQ3ZPAGTGPGMMV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"From Gradient Clipping to Normalization for Heavy Tailed SGD","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Florian H\\\"ubler, Ilyas Fatkhullin, Niao He","submitted_at":"2024-10-17T17:59:01Z","abstract_excerpt":"Recent empirical evidence indicates that many machine learning applications involve heavy-tailed gradient noise, which challenges the standard assumptions of bounded variance in stochastic optimization. Gradient clipping has emerged as a popular tool to handle this heavy-tailed noise, as it achieves good performance in this setting both theoretically and practically. However, our current theoretical understanding of non-convex gradient clipping has three main shortcomings. First, the theory hinges on large, increasing clipping thresholds, which are in stark contrast to the small constant clipp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13849","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/2410.13849/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":"2026-07-05T11:34:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3cYeAXembFe82RN8+SjhtU2CxP5i9X6Le3+EPvzvjHFLbhal84QjSNnSErgUUj7WaKyzCJSG6a+hO3x/jNS9CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T20:53:22.570745Z"},"content_sha256":"34802a0435bd099d732f8704e9454ba35d1d6426036e406f3c83bd98a66bf597","schema_version":"1.0","event_id":"sha256:34802a0435bd099d732f8704e9454ba35d1d6426036e406f3c83bd98a66bf597"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CTADMPSUWHKYMQ3ZPAGTGPGMMV/bundle.json","state_url":"https://pith.science/pith/CTADMPSUWHKYMQ3ZPAGTGPGMMV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CTADMPSUWHKYMQ3ZPAGTGPGMMV/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":"2026-08-08T20:53:22Z","links":{"resolver":"https://pith.science/pith/CTADMPSUWHKYMQ3ZPAGTGPGMMV","bundle":"https://pith.science/pith/CTADMPSUWHKYMQ3ZPAGTGPGMMV/bundle.json","state":"https://pith.science/pith/CTADMPSUWHKYMQ3ZPAGTGPGMMV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CTADMPSUWHKYMQ3ZPAGTGPGMMV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CTADMPSUWHKYMQ3ZPAGTGPGMMV","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":"9915a3a2caf3a444dd149a57793e13d2585cd1aec3bb11ab631a412b997433c2","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2024-10-17T17:59:01Z","title_canon_sha256":"c9f89eedac83b4f31f9921cf092f8cceb57b1c0699e5f7c58860d4d6ebb06248"},"schema_version":"1.0","source":{"id":"2410.13849","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.13849","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"arxiv_version","alias_value":"2410.13849v3","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13849","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"pith_short_12","alias_value":"CTADMPSUWHKY","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"pith_short_16","alias_value":"CTADMPSUWHKYMQ3Z","created_at":"2026-07-05T11:34:00Z"},{"alias_kind":"pith_short_8","alias_value":"CTADMPSU","created_at":"2026-07-05T11:34:00Z"}],"graph_snapshots":[{"event_id":"sha256:34802a0435bd099d732f8704e9454ba35d1d6426036e406f3c83bd98a66bf597","target":"graph","created_at":"2026-07-05T11:34:00Z","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/2410.13849/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent empirical evidence indicates that many machine learning applications involve heavy-tailed gradient noise, which challenges the standard assumptions of bounded variance in stochastic optimization. Gradient clipping has emerged as a popular tool to handle this heavy-tailed noise, as it achieves good performance in this setting both theoretically and practically. However, our current theoretical understanding of non-convex gradient clipping has three main shortcomings. First, the theory hinges on large, increasing clipping thresholds, which are in stark contrast to the small constant clipp","authors_text":"Florian H\\\"ubler, Ilyas Fatkhullin, Niao He","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2024-10-17T17:59:01Z","title":"From Gradient Clipping to Normalization for Heavy Tailed SGD"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13849","kind":"arxiv","version":3},"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:f139168830d7ea36181155061bb792755c6daefb6a13a87a58fc9ae440466b70","target":"record","created_at":"2026-07-05T11:34:00Z","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":"9915a3a2caf3a444dd149a57793e13d2585cd1aec3bb11ab631a412b997433c2","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2024-10-17T17:59:01Z","title_canon_sha256":"c9f89eedac83b4f31f9921cf092f8cceb57b1c0699e5f7c58860d4d6ebb06248"},"schema_version":"1.0","source":{"id":"2410.13849","kind":"arxiv","version":3}},"canonical_sha256":"14c0363e54b1d5864379780d333ccc6553b1968a306211af177db9981d6dfe3e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"14c0363e54b1d5864379780d333ccc6553b1968a306211af177db9981d6dfe3e","first_computed_at":"2026-07-05T11:34:00.672793Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:34:00.672793Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ogcx6xzaYZ3xz+eZd0wrDipGEp0d1qsQKvyUJEiiNBLEOPuKTH1w1HNdGQA25jkZXgzBdYmXSUu8T7nSSsLZCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:34:00.673420Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.13849","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f139168830d7ea36181155061bb792755c6daefb6a13a87a58fc9ae440466b70","sha256:34802a0435bd099d732f8704e9454ba35d1d6426036e406f3c83bd98a66bf597"],"state_sha256":"c2050b06261e9ba01d047f670468d76e470d050d3a515d3a8757a3bf9dd71304"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mC7KTSUajnmOX5tMNG9yE6/P+pOPUUGtVrelVFS+GbBl/ViEES982fmnPM5HSgK8cTQz6Nh8OD6f8+Gv1mqoBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T20:53:22.575384Z","bundle_sha256":"19fec0dd921b10234741dd9b6d3a3e2e46562f525be2fcd3b438d073e621f896"}}