{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:K6SCIRY4DY53YAW3TTEEI2DAUM","short_pith_number":"pith:K6SCIRY4","canonical_record":{"source":{"id":"2502.14123","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T21:55:11Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"fb022f1063b21dc246a5cfa5b43e4069855125a007f5f88ab676fdfa87f023dc","abstract_canon_sha256":"c8c8d9e55b1c09b0937aa572d67656305b0020c1117f59cc2b7e328ba7f21071"},"schema_version":"1.0"},"canonical_sha256":"57a424471c1e3bbc02db9cc8446860a3365ba3c67438b055ac2c18670a559edf","source":{"kind":"arxiv","id":"2502.14123","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.14123","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2502.14123v1","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14123","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"K6SCIRY4DY53","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"K6SCIRY4DY53YAW3","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"K6SCIRY4","created_at":"2026-07-05T10:17:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:K6SCIRY4DY53YAW3TTEEI2DAUM","target":"record","payload":{"canonical_record":{"source":{"id":"2502.14123","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T21:55:11Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"fb022f1063b21dc246a5cfa5b43e4069855125a007f5f88ab676fdfa87f023dc","abstract_canon_sha256":"c8c8d9e55b1c09b0937aa572d67656305b0020c1117f59cc2b7e328ba7f21071"},"schema_version":"1.0"},"canonical_sha256":"57a424471c1e3bbc02db9cc8446860a3365ba3c67438b055ac2c18670a559edf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:17:20.996262Z","signature_b64":"lV2NTdHatQoQpiKpxX7stXIJTtYsZrgXJYFw9e7Tzs+wsutmjQXOA+3G+PJ9scGmJOVF01m4OVmxb9hoRjXEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"57a424471c1e3bbc02db9cc8446860a3365ba3c67438b055ac2c18670a559edf","last_reissued_at":"2026-07-05T10:17:20.995795Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:17:20.995795Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.14123","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":"2026-07-05T10:17:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IllPDx0h0mkW1vvoTtHOoYkozdfzI86aQaFKqlOPr9fpVtmavOLYyDvRG8SCC4PxhhpgZue8kgfmf7XxhnOBDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:19:10.092476Z"},"content_sha256":"13e037c8a466b13aea153fa783cfab47911be5da5174937b022ac75020596c79","schema_version":"1.0","event_id":"sha256:13e037c8a466b13aea153fa783cfab47911be5da5174937b022ac75020596c79"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:K6SCIRY4DY53YAW3TTEEI2DAUM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Understanding SGD with Exponential Moving Average: A Case Study in Linear Regression","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Quanquan Gu, Xuheng Li","submitted_at":"2025-02-19T21:55:11Z","abstract_excerpt":"Exponential moving average (EMA) has recently gained significant popularity in training modern deep learning models, especially diffusion-based generative models. However, there have been few theoretical results explaining the effectiveness of EMA. In this paper, to better understand EMA, we establish the risk bound of online SGD with EMA for high-dimensional linear regression, one of the simplest overparameterized learning tasks that shares similarities with neural networks. Our results indicate that (i) the variance error of SGD with EMA is always smaller than that of SGD without averaging, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14123","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/2502.14123/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":"2026-07-05T10:17:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zCibSZlq9+Fd3ZVngsytVor4sS+jByKH+AfjSJhE8QFp9cL+VkoF76rTkkQUHPaKqR400opVcuI0BUvpl4I8DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:19:10.093010Z"},"content_sha256":"f817e53c99409e32d94dda06ce1bb9cc17810e1d32ae6f8bfb34b74c10814915","schema_version":"1.0","event_id":"sha256:f817e53c99409e32d94dda06ce1bb9cc17810e1d32ae6f8bfb34b74c10814915"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/K6SCIRY4DY53YAW3TTEEI2DAUM/bundle.json","state_url":"https://pith.science/pith/K6SCIRY4DY53YAW3TTEEI2DAUM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/K6SCIRY4DY53YAW3TTEEI2DAUM/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":"2026-08-08T09:19:10Z","links":{"resolver":"https://pith.science/pith/K6SCIRY4DY53YAW3TTEEI2DAUM","bundle":"https://pith.science/pith/K6SCIRY4DY53YAW3TTEEI2DAUM/bundle.json","state":"https://pith.science/pith/K6SCIRY4DY53YAW3TTEEI2DAUM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/K6SCIRY4DY53YAW3TTEEI2DAUM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:K6SCIRY4DY53YAW3TTEEI2DAUM","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":"c8c8d9e55b1c09b0937aa572d67656305b0020c1117f59cc2b7e328ba7f21071","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T21:55:11Z","title_canon_sha256":"fb022f1063b21dc246a5cfa5b43e4069855125a007f5f88ab676fdfa87f023dc"},"schema_version":"1.0","source":{"id":"2502.14123","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.14123","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"arxiv_version","alias_value":"2502.14123v1","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14123","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"pith_short_12","alias_value":"K6SCIRY4DY53","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"pith_short_16","alias_value":"K6SCIRY4DY53YAW3","created_at":"2026-07-05T10:17:20Z"},{"alias_kind":"pith_short_8","alias_value":"K6SCIRY4","created_at":"2026-07-05T10:17:20Z"}],"graph_snapshots":[{"event_id":"sha256:f817e53c99409e32d94dda06ce1bb9cc17810e1d32ae6f8bfb34b74c10814915","target":"graph","created_at":"2026-07-05T10:17:20Z","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/2502.14123/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Exponential moving average (EMA) has recently gained significant popularity in training modern deep learning models, especially diffusion-based generative models. However, there have been few theoretical results explaining the effectiveness of EMA. In this paper, to better understand EMA, we establish the risk bound of online SGD with EMA for high-dimensional linear regression, one of the simplest overparameterized learning tasks that shares similarities with neural networks. Our results indicate that (i) the variance error of SGD with EMA is always smaller than that of SGD without averaging, ","authors_text":"Quanquan Gu, Xuheng Li","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T21:55:11Z","title":"Understanding SGD with Exponential Moving Average: A Case Study in Linear Regression"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14123","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":"sha256:13e037c8a466b13aea153fa783cfab47911be5da5174937b022ac75020596c79","target":"record","created_at":"2026-07-05T10:17:20Z","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":"c8c8d9e55b1c09b0937aa572d67656305b0020c1117f59cc2b7e328ba7f21071","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-19T21:55:11Z","title_canon_sha256":"fb022f1063b21dc246a5cfa5b43e4069855125a007f5f88ab676fdfa87f023dc"},"schema_version":"1.0","source":{"id":"2502.14123","kind":"arxiv","version":1}},"canonical_sha256":"57a424471c1e3bbc02db9cc8446860a3365ba3c67438b055ac2c18670a559edf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"57a424471c1e3bbc02db9cc8446860a3365ba3c67438b055ac2c18670a559edf","first_computed_at":"2026-07-05T10:17:20.995795Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:17:20.995795Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lV2NTdHatQoQpiKpxX7stXIJTtYsZrgXJYFw9e7Tzs+wsutmjQXOA+3G+PJ9scGmJOVF01m4OVmxb9hoRjXEDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:17:20.996262Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.14123","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:13e037c8a466b13aea153fa783cfab47911be5da5174937b022ac75020596c79","sha256:f817e53c99409e32d94dda06ce1bb9cc17810e1d32ae6f8bfb34b74c10814915"],"state_sha256":"80c194fac9d781ff8148e65311d6dab5246ae0774c8efcc6735983eb6c9d3f50"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NZl4K2rUfRcG/TY71qPkTc2RW2tojz8s+XFDJ042a0cnrg/G18HrIC+A6Kn5foUq9FFAcFdI8uHW9VlcemDtDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T09:19:10.096060Z","bundle_sha256":"8c7e0bb536c906bbec2d2c280a7f68ba0027fc0d3bc8dbd7bcdb9b958abebb1d"}}