{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:TY6IXUFLDYRT3A3CE6T3F4GS3A","short_pith_number":"pith:TY6IXUFL","schema_version":"1.0","canonical_sha256":"9e3c8bd0ab1e233d836227a7b2f0d2d83026abe985e5a245425d7233414d59f9","source":{"kind":"arxiv","id":"1608.08225","version":4},"attestation_state":"computed","paper":{"title":"Why does deep and cheap learning work so well?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE","stat.ML"],"primary_cat":"cond-mat.dis-nn","authors_text":"David Rolnick (MIT), Henry W. Lin (Harvard), Max Tegmark (MIT)","submitted_at":"2016-08-29T20:00:14Z","abstract_excerpt":"We show how the success of deep learning could depend not only on mathematics but also on physics: although well-known mathematical theorems guarantee that neural networks can approximate arbitrary functions well, the class of functions of practical interest can frequently be approximated through \"cheap learning\" with exponentially fewer parameters than generic ones. We explore how properties frequently encountered in physics such as symmetry, locality, compositionality, and polynomial log-probability translate into exceptionally simple neural networks. We further argue that when the statistic"},"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":"1608.08225","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cond-mat.dis-nn","submitted_at":"2016-08-29T20:00:14Z","cross_cats_sorted":["cs.LG","cs.NE","stat.ML"],"title_canon_sha256":"13ae7ada59db9998f74c9bfb421e0300784bbfb483ec9d1d900c92a2234cd833","abstract_canon_sha256":"84079cfcd1012e8dfb22ec4d0409cfb78312f5a437720907c04c48ac9bd02a06"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:35:24.765128Z","signature_b64":"n75kVa1yeG67JCqdwf1OXTu7luyJkRw2RRU2OfLYY1DsaRM4cgGgwZ1FVebSxVPoxjMAOgIusa94aH77Kru2Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9e3c8bd0ab1e233d836227a7b2f0d2d83026abe985e5a245425d7233414d59f9","last_reissued_at":"2026-05-18T00:35:24.764758Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:35:24.764758Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why does deep and cheap learning work so well?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE","stat.ML"],"primary_cat":"cond-mat.dis-nn","authors_text":"David Rolnick (MIT), Henry W. Lin (Harvard), Max Tegmark (MIT)","submitted_at":"2016-08-29T20:00:14Z","abstract_excerpt":"We show how the success of deep learning could depend not only on mathematics but also on physics: although well-known mathematical theorems guarantee that neural networks can approximate arbitrary functions well, the class of functions of practical interest can frequently be approximated through \"cheap learning\" with exponentially fewer parameters than generic ones. We explore how properties frequently encountered in physics such as symmetry, locality, compositionality, and polynomial log-probability translate into exceptionally simple neural networks. We further argue that when the statistic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1608.08225","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"1608.08225","created_at":"2026-05-18T00:35:24.764817+00:00"},{"alias_kind":"arxiv_version","alias_value":"1608.08225v4","created_at":"2026-05-18T00:35:24.764817+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1608.08225","created_at":"2026-05-18T00:35:24.764817+00:00"},{"alias_kind":"pith_short_12","alias_value":"TY6IXUFLDYRT","created_at":"2026-05-18T12:30:46.583412+00:00"},{"alias_kind":"pith_short_16","alias_value":"TY6IXUFLDYRT3A3C","created_at":"2026-05-18T12:30:46.583412+00:00"},{"alias_kind":"pith_short_8","alias_value":"TY6IXUFL","created_at":"2026-05-18T12:30:46.583412+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.00333","citing_title":"Borrowed Geometry: Cross-Distribution Head-Importance Fingerprints of Frozen Pretrained Gemma 4 31B","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.00333","citing_title":"Borrowed Geometry: Cross-Distribution Head-Importance Fingerprints of Frozen Pretrained Gemma 4 31B","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A","json":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A.json","graph_json":"https://pith.science/api/pith-number/TY6IXUFLDYRT3A3CE6T3F4GS3A/graph.json","events_json":"https://pith.science/api/pith-number/TY6IXUFLDYRT3A3CE6T3F4GS3A/events.json","paper":"https://pith.science/paper/TY6IXUFL"},"agent_actions":{"view_html":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A","download_json":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A.json","view_paper":"https://pith.science/paper/TY6IXUFL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1608.08225&json=true","fetch_graph":"https://pith.science/api/pith-number/TY6IXUFLDYRT3A3CE6T3F4GS3A/graph.json","fetch_events":"https://pith.science/api/pith-number/TY6IXUFLDYRT3A3CE6T3F4GS3A/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A/action/storage_attestation","attest_author":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A/action/author_attestation","sign_citation":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A/action/citation_signature","submit_replication":"https://pith.science/pith/TY6IXUFLDYRT3A3CE6T3F4GS3A/action/replication_record"}},"created_at":"2026-05-18T00:35:24.764817+00:00","updated_at":"2026-05-18T00:35:24.764817+00:00"}