{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:E2BBXRZLMQC6GJGX4BC7A7C6AZ","short_pith_number":"pith:E2BBXRZL","schema_version":"1.0","canonical_sha256":"26821bc72b6405e324d7e045f07c5e06604c3c857f59b695e2fe9f2abe6044be","source":{"kind":"arxiv","id":"2306.13914","version":1},"attestation_state":"computed","paper":{"title":"G-TRACER: Expected Sharpness Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"John Williams, Stephen Roberts","submitted_at":"2023-06-24T09:28:49Z","abstract_excerpt":"We propose a new regularization scheme for the optimization of deep learning architectures, G-TRACER (\"Geometric TRACE Ratio\"), which promotes generalization by seeking flat minima, and has a sound theoretical basis as an approximation to a natural-gradient descent based optimization of a generalized Bayes objective. By augmenting the loss function with a TRACER, curvature-regularized optimizers (eg SGD-TRACER and Adam-TRACER) are simple to implement as modifications to existing optimizers and don't require extensive tuning. We show that the method converges to a neighborhood (depending on the"},"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":"2306.13914","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-06-24T09:28:49Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"125d068e1ecfefbf72aacda9a90c2c1e4fdfb277bfabd7fab9461536758f39e1","abstract_canon_sha256":"359b4bc52cdbd2af86079fa55777840f524f5822261064dfffb44fd08ec3daee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:24:17.550683Z","signature_b64":"9jjcMydn48SYEJ7UZf3w2lXFRRX7yeRPHv4DQpYroiZjjZCk2K8SqRgBNX2ANX903Q0W/k/VuX2YUD8sPoOQAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"26821bc72b6405e324d7e045f07c5e06604c3c857f59b695e2fe9f2abe6044be","last_reissued_at":"2026-07-05T06:24:17.550285Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:24:17.550285Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"G-TRACER: Expected Sharpness Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"John Williams, Stephen Roberts","submitted_at":"2023-06-24T09:28:49Z","abstract_excerpt":"We propose a new regularization scheme for the optimization of deep learning architectures, G-TRACER (\"Geometric TRACE Ratio\"), which promotes generalization by seeking flat minima, and has a sound theoretical basis as an approximation to a natural-gradient descent based optimization of a generalized Bayes objective. By augmenting the loss function with a TRACER, curvature-regularized optimizers (eg SGD-TRACER and Adam-TRACER) are simple to implement as modifications to existing optimizers and don't require extensive tuning. We show that the method converges to a neighborhood (depending on the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.13914","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/2306.13914/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":"2306.13914","created_at":"2026-07-05T06:24:17.550342+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.13914v1","created_at":"2026-07-05T06:24:17.550342+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.13914","created_at":"2026-07-05T06:24:17.550342+00:00"},{"alias_kind":"pith_short_12","alias_value":"E2BBXRZLMQC6","created_at":"2026-07-05T06:24:17.550342+00:00"},{"alias_kind":"pith_short_16","alias_value":"E2BBXRZLMQC6GJGX","created_at":"2026-07-05T06:24:17.550342+00:00"},{"alias_kind":"pith_short_8","alias_value":"E2BBXRZL","created_at":"2026-07-05T06:24:17.550342+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ","json":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ.json","graph_json":"https://pith.science/api/pith-number/E2BBXRZLMQC6GJGX4BC7A7C6AZ/graph.json","events_json":"https://pith.science/api/pith-number/E2BBXRZLMQC6GJGX4BC7A7C6AZ/events.json","paper":"https://pith.science/paper/E2BBXRZL"},"agent_actions":{"view_html":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ","download_json":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ.json","view_paper":"https://pith.science/paper/E2BBXRZL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.13914&json=true","fetch_graph":"https://pith.science/api/pith-number/E2BBXRZLMQC6GJGX4BC7A7C6AZ/graph.json","fetch_events":"https://pith.science/api/pith-number/E2BBXRZLMQC6GJGX4BC7A7C6AZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ/action/storage_attestation","attest_author":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ/action/author_attestation","sign_citation":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ/action/citation_signature","submit_replication":"https://pith.science/pith/E2BBXRZLMQC6GJGX4BC7A7C6AZ/action/replication_record"}},"created_at":"2026-07-05T06:24:17.550342+00:00","updated_at":"2026-07-05T06:24:17.550342+00:00"}