{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:VHLBLFQX2N2LP7K7LJRW6LMQ36","short_pith_number":"pith:VHLBLFQX","schema_version":"1.0","canonical_sha256":"a9d6159617d374b7fd5f5a636f2d90df9d4b57edca5417277b24bcfe004837f9","source":{"kind":"arxiv","id":"2009.13094","version":1},"attestation_state":"computed","paper":{"title":"Improved generalization by noise enhancement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.dis-nn","stat.ML"],"primary_cat":"cs.LG","authors_text":"Masahito Ueda, Takashi Mori","submitted_at":"2020-09-28T06:29:23Z","abstract_excerpt":"Recent studies have demonstrated that noise in stochastic gradient descent (SGD) is closely related to generalization: A larger SGD noise, if not too large, results in better generalization. Since the covariance of the SGD noise is proportional to $\\eta^2/B$, where $\\eta$ is the learning rate and $B$ is the minibatch size of SGD, the SGD noise has so far been controlled by changing $\\eta$ and/or $B$. However, too large $\\eta$ results in instability in the training dynamics and a small $B$ prevents scalable parallel computation. It is thus desirable to develop a method of controlling the SGD no"},"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":"2009.13094","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-09-28T06:29:23Z","cross_cats_sorted":["cond-mat.dis-nn","stat.ML"],"title_canon_sha256":"2a758c054fd7be19b861fdf93da13ed1cb72eab648c3999ad002598af10cf956","abstract_canon_sha256":"1357af6318d11d4f7fb90f99f87ed79be5f6aa44df365ccf2d8ebd76d73368b1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:38:26.526187Z","signature_b64":"VkTetSvVNS870gI6rgqvEQjxN7tuBqZyI3P/MYt5anCB2Bv9OLu4RwMAuxxoHzJqU0w1V3o0pEmuHzlQOjhKCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a9d6159617d374b7fd5f5a636f2d90df9d4b57edca5417277b24bcfe004837f9","last_reissued_at":"2026-07-05T01:38:26.525631Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:38:26.525631Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improved generalization by noise enhancement","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.dis-nn","stat.ML"],"primary_cat":"cs.LG","authors_text":"Masahito Ueda, Takashi Mori","submitted_at":"2020-09-28T06:29:23Z","abstract_excerpt":"Recent studies have demonstrated that noise in stochastic gradient descent (SGD) is closely related to generalization: A larger SGD noise, if not too large, results in better generalization. Since the covariance of the SGD noise is proportional to $\\eta^2/B$, where $\\eta$ is the learning rate and $B$ is the minibatch size of SGD, the SGD noise has so far been controlled by changing $\\eta$ and/or $B$. However, too large $\\eta$ results in instability in the training dynamics and a small $B$ prevents scalable parallel computation. It is thus desirable to develop a method of controlling the SGD no"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.13094","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/2009.13094/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":"2009.13094","created_at":"2026-07-05T01:38:26.525695+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.13094v1","created_at":"2026-07-05T01:38:26.525695+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.13094","created_at":"2026-07-05T01:38:26.525695+00:00"},{"alias_kind":"pith_short_12","alias_value":"VHLBLFQX2N2L","created_at":"2026-07-05T01:38:26.525695+00:00"},{"alias_kind":"pith_short_16","alias_value":"VHLBLFQX2N2LP7K7","created_at":"2026-07-05T01:38:26.525695+00:00"},{"alias_kind":"pith_short_8","alias_value":"VHLBLFQX","created_at":"2026-07-05T01:38:26.525695+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00293","citing_title":"Accurate Large-sample Uncertainty Quantification using Stochastic Gradient Markov Chain Monte Carlo","ref_index":50,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36","json":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36.json","graph_json":"https://pith.science/api/pith-number/VHLBLFQX2N2LP7K7LJRW6LMQ36/graph.json","events_json":"https://pith.science/api/pith-number/VHLBLFQX2N2LP7K7LJRW6LMQ36/events.json","paper":"https://pith.science/paper/VHLBLFQX"},"agent_actions":{"view_html":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36","download_json":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36.json","view_paper":"https://pith.science/paper/VHLBLFQX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.13094&json=true","fetch_graph":"https://pith.science/api/pith-number/VHLBLFQX2N2LP7K7LJRW6LMQ36/graph.json","fetch_events":"https://pith.science/api/pith-number/VHLBLFQX2N2LP7K7LJRW6LMQ36/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36/action/storage_attestation","attest_author":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36/action/author_attestation","sign_citation":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36/action/citation_signature","submit_replication":"https://pith.science/pith/VHLBLFQX2N2LP7K7LJRW6LMQ36/action/replication_record"}},"created_at":"2026-07-05T01:38:26.525695+00:00","updated_at":"2026-07-05T01:38:26.525695+00:00"}