{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2016:YKNN26FL5XDXD7YXENFDXI3T27","short_pith_number":"pith:YKNN26FL","schema_version":"1.0","canonical_sha256":"c29add78abedc771ff17234a3ba373d7c9aa28c26457669d3da4f7a19aefb6e1","source":{"kind":"arxiv","id":"1606.05340","version":2},"attestation_state":"computed","paper":{"title":"Exponential expressivity in deep neural networks through transient chaos","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.LG"],"primary_cat":"stat.ML","authors_text":"Ben Poole, Jascha Sohl-Dickstein, Maithra Raghu, Subhaneil Lahiri, Surya Ganguli","submitted_at":"2016-06-16T19:59:57Z","abstract_excerpt":"We combine Riemannian geometry with the mean field theory of high dimensional chaos to study the nature of signal propagation in generic, deep neural networks with random weights. Our results reveal an order-to-chaos expressivity phase transition, with networks in the chaotic phase computing nonlinear functions whose global curvature grows exponentially with depth but not width. We prove this generic class of deep random functions cannot be efficiently computed by any shallow network, going beyond prior work restricted to the analysis of single functions. Moreover, we formalize and quantitativ"},"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":"1606.05340","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2016-06-16T19:59:57Z","cross_cats_sorted":["cond-mat.dis-nn","cs.LG"],"title_canon_sha256":"db6eade9e497ca57b1167d74ae260b9066d61e1c5ef03607c4df1a416d04b6e9","abstract_canon_sha256":"ada722806730540472d186a8b49f05a6a7899866b79685891b27a677f89e470e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:12:11.321812Z","signature_b64":"EfaZw7gnEFUA81BkvQrXkMxOHhO4r1q35UBYdamAcFu3+IXzlpd5CvMzvH0AlAxM/jpNJskdlJBU2U3dVYeOAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c29add78abedc771ff17234a3ba373d7c9aa28c26457669d3da4f7a19aefb6e1","last_reissued_at":"2026-05-18T01:12:11.321470Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:12:11.321470Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Exponential expressivity in deep neural networks through transient chaos","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.dis-nn","cs.LG"],"primary_cat":"stat.ML","authors_text":"Ben Poole, Jascha Sohl-Dickstein, Maithra Raghu, Subhaneil Lahiri, Surya Ganguli","submitted_at":"2016-06-16T19:59:57Z","abstract_excerpt":"We combine Riemannian geometry with the mean field theory of high dimensional chaos to study the nature of signal propagation in generic, deep neural networks with random weights. Our results reveal an order-to-chaos expressivity phase transition, with networks in the chaotic phase computing nonlinear functions whose global curvature grows exponentially with depth but not width. We prove this generic class of deep random functions cannot be efficiently computed by any shallow network, going beyond prior work restricted to the analysis of single functions. Moreover, we formalize and quantitativ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1606.05340","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":""},"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":"1606.05340","created_at":"2026-05-18T01:12:11.321524+00:00"},{"alias_kind":"arxiv_version","alias_value":"1606.05340v2","created_at":"2026-05-18T01:12:11.321524+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1606.05340","created_at":"2026-05-18T01:12:11.321524+00:00"},{"alias_kind":"pith_short_12","alias_value":"YKNN26FL5XDX","created_at":"2026-05-18T12:30:53.716459+00:00"},{"alias_kind":"pith_short_16","alias_value":"YKNN26FL5XDXD7YX","created_at":"2026-05-18T12:30:53.716459+00:00"},{"alias_kind":"pith_short_8","alias_value":"YKNN26FL","created_at":"2026-05-18T12:30:53.716459+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2605.30860","citing_title":"Bayesian Inference with Shaped Deep Non-linear MLPs","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12536","citing_title":"Information as Maximum-Caliber Deviation: A bridge between Integrated Information Theory and the Free Energy Principle","ref_index":152,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27","json":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27.json","graph_json":"https://pith.science/api/pith-number/YKNN26FL5XDXD7YXENFDXI3T27/graph.json","events_json":"https://pith.science/api/pith-number/YKNN26FL5XDXD7YXENFDXI3T27/events.json","paper":"https://pith.science/paper/YKNN26FL"},"agent_actions":{"view_html":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27","download_json":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27.json","view_paper":"https://pith.science/paper/YKNN26FL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1606.05340&json=true","fetch_graph":"https://pith.science/api/pith-number/YKNN26FL5XDXD7YXENFDXI3T27/graph.json","fetch_events":"https://pith.science/api/pith-number/YKNN26FL5XDXD7YXENFDXI3T27/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27/action/storage_attestation","attest_author":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27/action/author_attestation","sign_citation":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27/action/citation_signature","submit_replication":"https://pith.science/pith/YKNN26FL5XDXD7YXENFDXI3T27/action/replication_record"}},"created_at":"2026-05-18T01:12:11.321524+00:00","updated_at":"2026-05-18T01:12:11.321524+00:00"}