{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:BYOYC5GRP6DHMB6BBRAQ2IPEYX","short_pith_number":"pith:BYOYC5GR","schema_version":"1.0","canonical_sha256":"0e1d8174d17f867607c10c410d21e4c5c1fe77acbdd2b28820e0cc779a76a29f","source":{"kind":"arxiv","id":"2209.13083","version":2},"attestation_state":"computed","paper":{"title":"Why neural networks find simple solutions: the many regularizers of geometric complexity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Benoit Dherin, David G.T. Barrett, Michael Munn, Mihaela Rosca","submitted_at":"2022-09-27T00:16:38Z","abstract_excerpt":"In many contexts, simpler models are preferable to more complex models and the control of this model complexity is the goal for many methods in machine learning such as regularization, hyperparameter tuning and architecture design. In deep learning, it has been difficult to understand the underlying mechanisms of complexity control, since many traditional measures are not naturally suitable for deep neural networks. Here we develop the notion of geometric complexity, which is a measure of the variability of the model function, computed using a discrete Dirichlet energy. Using a combination of "},"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":"2209.13083","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-09-27T00:16:38Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"5405b113933c13570515affe917b1997533168dee5cad779054ca9bfdc9bf6d6","abstract_canon_sha256":"2cbd858949485b2e71d23f43c23c205d461e91313d50638225292e18a1ec0cfd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:28:09.198938Z","signature_b64":"FpWtd+naU1K4Ytu3Y+ul41TwO58buMP+Hn/69h8PCLpkEMBGynoVbotwx5Ip3ZNIqwZepzUXJKwXxlSc1xbtBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0e1d8174d17f867607c10c410d21e4c5c1fe77acbdd2b28820e0cc779a76a29f","last_reissued_at":"2026-07-05T05:28:09.198600Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:28:09.198600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why neural networks find simple solutions: the many regularizers of geometric complexity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Benoit Dherin, David G.T. Barrett, Michael Munn, Mihaela Rosca","submitted_at":"2022-09-27T00:16:38Z","abstract_excerpt":"In many contexts, simpler models are preferable to more complex models and the control of this model complexity is the goal for many methods in machine learning such as regularization, hyperparameter tuning and architecture design. In deep learning, it has been difficult to understand the underlying mechanisms of complexity control, since many traditional measures are not naturally suitable for deep neural networks. Here we develop the notion of geometric complexity, which is a measure of the variability of the model function, computed using a discrete Dirichlet energy. Using a combination of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.13083","kind":"arxiv","version":2},"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/2209.13083/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":"2209.13083","created_at":"2026-07-05T05:28:09.198656+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.13083v2","created_at":"2026-07-05T05:28:09.198656+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.13083","created_at":"2026-07-05T05:28:09.198656+00:00"},{"alias_kind":"pith_short_12","alias_value":"BYOYC5GRP6DH","created_at":"2026-07-05T05:28:09.198656+00:00"},{"alias_kind":"pith_short_16","alias_value":"BYOYC5GRP6DHMB6B","created_at":"2026-07-05T05:28:09.198656+00:00"},{"alias_kind":"pith_short_8","alias_value":"BYOYC5GR","created_at":"2026-07-05T05:28:09.198656+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/BYOYC5GRP6DHMB6BBRAQ2IPEYX","json":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX.json","graph_json":"https://pith.science/api/pith-number/BYOYC5GRP6DHMB6BBRAQ2IPEYX/graph.json","events_json":"https://pith.science/api/pith-number/BYOYC5GRP6DHMB6BBRAQ2IPEYX/events.json","paper":"https://pith.science/paper/BYOYC5GR"},"agent_actions":{"view_html":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX","download_json":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX.json","view_paper":"https://pith.science/paper/BYOYC5GR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.13083&json=true","fetch_graph":"https://pith.science/api/pith-number/BYOYC5GRP6DHMB6BBRAQ2IPEYX/graph.json","fetch_events":"https://pith.science/api/pith-number/BYOYC5GRP6DHMB6BBRAQ2IPEYX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX/action/storage_attestation","attest_author":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX/action/author_attestation","sign_citation":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX/action/citation_signature","submit_replication":"https://pith.science/pith/BYOYC5GRP6DHMB6BBRAQ2IPEYX/action/replication_record"}},"created_at":"2026-07-05T05:28:09.198656+00:00","updated_at":"2026-07-05T05:28:09.198656+00:00"}