{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:NAFXKFONGEUJI7GP2NKZLUZJPQ","short_pith_number":"pith:NAFXKFON","schema_version":"1.0","canonical_sha256":"680b7515cd3128947ccfd35595d3297c0e9e9faf013f350e019605b306a4d666","source":{"kind":"arxiv","id":"1912.02292","version":1},"attestation_state":"computed","paper":{"title":"Deep Double Descent: Where Bigger Models and More Data Hurt","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Boaz Barak, Gal Kaplun, Ilya Sutskever, Preetum Nakkiran, Tristan Yang, Yamini Bansal","submitted_at":"2019-12-04T22:47:31Z","abstract_excerpt":"We show that a variety of modern deep learning tasks exhibit a \"double-descent\" phenomenon where, as we increase model size, performance first gets worse and then gets better. Moreover, we show that double descent occurs not just as a function of model size, but also as a function of the number of training epochs. We unify the above phenomena by defining a new complexity measure we call the effective model complexity and conjecture a generalized double descent with respect to this measure. Furthermore, our notion of model complexity allows us to identify certain regimes where increasing (even "},"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":"1912.02292","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-12-04T22:47:31Z","cross_cats_sorted":["cs.CV","cs.NE","stat.ML"],"title_canon_sha256":"5760da5f113a1020b9704671b696004dd10dfbe1e1499e93dbd9f938425fabb1","abstract_canon_sha256":"9b0259bc18125dc2f855106680d46d3702cb67f532e46dd056d403585ec5ef8d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:24:12.254670Z","signature_b64":"lHFNVG0TJuqPOjbiRs/g8feqEJhk95yi8jEo3Gbj6KdhBo1PlKMb/T6iXjoio2cvK+V4RBh6wPXDZKskKbqHDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"680b7515cd3128947ccfd35595d3297c0e9e9faf013f350e019605b306a4d666","last_reissued_at":"2026-07-05T00:24:12.254167Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:24:12.254167Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Double Descent: Where Bigger Models and More Data Hurt","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Boaz Barak, Gal Kaplun, Ilya Sutskever, Preetum Nakkiran, Tristan Yang, Yamini Bansal","submitted_at":"2019-12-04T22:47:31Z","abstract_excerpt":"We show that a variety of modern deep learning tasks exhibit a \"double-descent\" phenomenon where, as we increase model size, performance first gets worse and then gets better. Moreover, we show that double descent occurs not just as a function of model size, but also as a function of the number of training epochs. We unify the above phenomena by defining a new complexity measure we call the effective model complexity and conjecture a generalized double descent with respect to this measure. Furthermore, our notion of model complexity allows us to identify certain regimes where increasing (even "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1912.02292","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/1912.02292/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":"1912.02292","created_at":"2026-07-05T00:24:12.254225+00:00"},{"alias_kind":"arxiv_version","alias_value":"1912.02292v1","created_at":"2026-07-05T00:24:12.254225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1912.02292","created_at":"2026-07-05T00:24:12.254225+00:00"},{"alias_kind":"pith_short_12","alias_value":"NAFXKFONGEUJ","created_at":"2026-07-05T00:24:12.254225+00:00"},{"alias_kind":"pith_short_16","alias_value":"NAFXKFONGEUJI7GP","created_at":"2026-07-05T00:24:12.254225+00:00"},{"alias_kind":"pith_short_8","alias_value":"NAFXKFON","created_at":"2026-07-05T00:24:12.254225+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.13370","citing_title":"A Quantitative Experimental Repeated Measures Study of Training Dynamics in a Small Llama Style Language Model Under a Compute-Aware Token Budget","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26248","citing_title":"Unified Neural Scaling Laws","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07327","citing_title":"Six Open Questions in Machine-Learned Interatomic Potential Foundation Models","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23591","citing_title":"Asymmetric Scaling Laws from Sparse Features","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15253","citing_title":"Position: Ideas Should be the Center of Machine Learning Research","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2102.01293","citing_title":"Scaling Laws for Transfer","ref_index":181,"is_internal_anchor":false},{"citing_arxiv_id":"2205.10487","citing_title":"Scaling Laws and Interpretability of Learning from Repeated Data","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14200","citing_title":"How to Scale Mixture-of-Experts: From muP to the Maximally Scale-Stable Parameterization","ref_index":256,"is_internal_anchor":false},{"citing_arxiv_id":"2201.02177","citing_title":"Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2112.00861","citing_title":"A General Language Assistant as a Laboratory for Alignment","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2207.05221","citing_title":"Language Models (Mostly) Know What They Know","ref_index":94,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ","json":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ.json","graph_json":"https://pith.science/api/pith-number/NAFXKFONGEUJI7GP2NKZLUZJPQ/graph.json","events_json":"https://pith.science/api/pith-number/NAFXKFONGEUJI7GP2NKZLUZJPQ/events.json","paper":"https://pith.science/paper/NAFXKFON"},"agent_actions":{"view_html":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ","download_json":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ.json","view_paper":"https://pith.science/paper/NAFXKFON","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1912.02292&json=true","fetch_graph":"https://pith.science/api/pith-number/NAFXKFONGEUJI7GP2NKZLUZJPQ/graph.json","fetch_events":"https://pith.science/api/pith-number/NAFXKFONGEUJI7GP2NKZLUZJPQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ/action/storage_attestation","attest_author":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ/action/author_attestation","sign_citation":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ/action/citation_signature","submit_replication":"https://pith.science/pith/NAFXKFONGEUJI7GP2NKZLUZJPQ/action/replication_record"}},"created_at":"2026-07-05T00:24:12.254225+00:00","updated_at":"2026-07-05T00:24:12.254225+00:00"}