{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ACRR2QMPZXPVMSPUD4D2XLT5X4","short_pith_number":"pith:ACRR2QMP","schema_version":"1.0","canonical_sha256":"00a31d418fcddf5649f41f07abae7dbf17d05307e158e28471c892d839c609b2","source":{"kind":"arxiv","id":"2405.13098","version":4},"attestation_state":"computed","paper":{"title":"Glassy dynamics in deep neural networks: A structural comparison","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cond-mat.stat-mech"],"primary_cat":"physics.comp-ph","authors_text":"Liesbeth M. C. Janssen, Max Kerr Winter","submitted_at":"2024-05-21T14:43:02Z","abstract_excerpt":"Deep Neural Networks (DNNs) share important similarities with structural glasses. Both have many degrees of freedom, and their dynamics are governed by a high-dimensional, non-convex landscape representing either the loss or energy, respectively. Furthermore, both experience gradient descent dynamics subject to noise. In this work we investigate, by performing quantitative measurements on realistic networks trained on the MNIST and CIFAR-10 datasets, the extent to which this qualitative similarity gives rise to glass-like dynamics in neural networks. We demonstrate the existence of a Topology "},"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":"2405.13098","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"physics.comp-ph","submitted_at":"2024-05-21T14:43:02Z","cross_cats_sorted":["cond-mat.dis-nn","cond-mat.stat-mech"],"title_canon_sha256":"9213c56a647257ba7f352d56d5562d6179a193b307476c72175e058986a95cdc","abstract_canon_sha256":"ac34f1c3d9781d346d02ff2c8272bef9089932f65f801b386a936e385ca85e71"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:37:26.395190Z","signature_b64":"9dvnGQbUIGImlPY/SE3VFlDSmc+BbgV7Hu1nLFe4KipYlYoY6a/Fh86TZDExl96ID++tMSVbg7tpxQC/q676Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"00a31d418fcddf5649f41f07abae7dbf17d05307e158e28471c892d839c609b2","last_reissued_at":"2026-07-05T10:37:26.394574Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:37:26.394574Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Glassy dynamics in deep neural networks: A structural comparison","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn","cond-mat.stat-mech"],"primary_cat":"physics.comp-ph","authors_text":"Liesbeth M. C. Janssen, Max Kerr Winter","submitted_at":"2024-05-21T14:43:02Z","abstract_excerpt":"Deep Neural Networks (DNNs) share important similarities with structural glasses. Both have many degrees of freedom, and their dynamics are governed by a high-dimensional, non-convex landscape representing either the loss or energy, respectively. Furthermore, both experience gradient descent dynamics subject to noise. In this work we investigate, by performing quantitative measurements on realistic networks trained on the MNIST and CIFAR-10 datasets, the extent to which this qualitative similarity gives rise to glass-like dynamics in neural networks. We demonstrate the existence of a Topology "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.13098","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2405.13098/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":"2405.13098","created_at":"2026-07-05T10:37:26.394640+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.13098v4","created_at":"2026-07-05T10:37:26.394640+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.13098","created_at":"2026-07-05T10:37:26.394640+00:00"},{"alias_kind":"pith_short_12","alias_value":"ACRR2QMPZXPV","created_at":"2026-07-05T10:37:26.394640+00:00"},{"alias_kind":"pith_short_16","alias_value":"ACRR2QMPZXPVMSPU","created_at":"2026-07-05T10:37:26.394640+00:00"},{"alias_kind":"pith_short_8","alias_value":"ACRR2QMP","created_at":"2026-07-05T10:37:26.394640+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.06789","citing_title":"Liquid and solid layers in a thermal deep learning machine","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4","json":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4.json","graph_json":"https://pith.science/api/pith-number/ACRR2QMPZXPVMSPUD4D2XLT5X4/graph.json","events_json":"https://pith.science/api/pith-number/ACRR2QMPZXPVMSPUD4D2XLT5X4/events.json","paper":"https://pith.science/paper/ACRR2QMP"},"agent_actions":{"view_html":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4","download_json":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4.json","view_paper":"https://pith.science/paper/ACRR2QMP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.13098&json=true","fetch_graph":"https://pith.science/api/pith-number/ACRR2QMPZXPVMSPUD4D2XLT5X4/graph.json","fetch_events":"https://pith.science/api/pith-number/ACRR2QMPZXPVMSPUD4D2XLT5X4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4/action/storage_attestation","attest_author":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4/action/author_attestation","sign_citation":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4/action/citation_signature","submit_replication":"https://pith.science/pith/ACRR2QMPZXPVMSPUD4D2XLT5X4/action/replication_record"}},"created_at":"2026-07-05T10:37:26.394640+00:00","updated_at":"2026-07-05T10:37:26.394640+00:00"}