{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:CKVTORSE5P2MLK5JYKB3KSNRKD","short_pith_number":"pith:CKVTORSE","schema_version":"1.0","canonical_sha256":"12ab374644ebf4c5aba9c283b549b150f85c6b512a87f7da5f629d1ea69a2493","source":{"kind":"arxiv","id":"2306.03521","version":1},"attestation_state":"computed","paper":{"title":"Machine learning in and out of equilibrium","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cond-mat.stat-mech"],"primary_cat":"cs.LG","authors_text":"Alexander Strang, Alkan Kabak\\c{c}{\\i}o\\u{g}lu, Deniz Yuret, Michael Hinczewski, Shishir Adhikari","submitted_at":"2023-06-06T09:12:49Z","abstract_excerpt":"The algorithms used to train neural networks, like stochastic gradient descent (SGD), have close parallels to natural processes that navigate a high-dimensional parameter space -- for example protein folding or evolution. Our study uses a Fokker-Planck approach, adapted from statistical physics, to explore these parallels in a single, unified framework. We focus in particular on the stationary state of the system in the long-time limit, which in conventional SGD is out of equilibrium, exhibiting persistent currents in the space of network parameters. As in its physical analogues, the current i"},"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":"2306.03521","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-06T09:12:49Z","cross_cats_sorted":["cond-mat.stat-mech"],"title_canon_sha256":"4d5c9b28671bc359d209cc9b2f7132f4813e7bde6e485c21dd858c29e6344d85","abstract_canon_sha256":"a333a1ba95d2b9bd3f161b222694b69434e8b73c77888d1b646e6e3173dae2b6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:17:59.049120Z","signature_b64":"NSbYl0IruBZci5jUlpmZh/6nbQGM7shjoa/RVvGaP0/PH8xsPUTnod9wm1hY77zH6mmA17yjwx4Z5R2RyjMsBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"12ab374644ebf4c5aba9c283b549b150f85c6b512a87f7da5f629d1ea69a2493","last_reissued_at":"2026-07-05T06:17:59.048660Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:17:59.048660Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Machine learning in and out of equilibrium","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cond-mat.stat-mech"],"primary_cat":"cs.LG","authors_text":"Alexander Strang, Alkan Kabak\\c{c}{\\i}o\\u{g}lu, Deniz Yuret, Michael Hinczewski, Shishir Adhikari","submitted_at":"2023-06-06T09:12:49Z","abstract_excerpt":"The algorithms used to train neural networks, like stochastic gradient descent (SGD), have close parallels to natural processes that navigate a high-dimensional parameter space -- for example protein folding or evolution. Our study uses a Fokker-Planck approach, adapted from statistical physics, to explore these parallels in a single, unified framework. We focus in particular on the stationary state of the system in the long-time limit, which in conventional SGD is out of equilibrium, exhibiting persistent currents in the space of network parameters. As in its physical analogues, the current i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.03521","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/2306.03521/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":"2306.03521","created_at":"2026-07-05T06:17:59.048717+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.03521v1","created_at":"2026-07-05T06:17:59.048717+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.03521","created_at":"2026-07-05T06:17:59.048717+00:00"},{"alias_kind":"pith_short_12","alias_value":"CKVTORSE5P2M","created_at":"2026-07-05T06:17:59.048717+00:00"},{"alias_kind":"pith_short_16","alias_value":"CKVTORSE5P2MLK5J","created_at":"2026-07-05T06:17:59.048717+00:00"},{"alias_kind":"pith_short_8","alias_value":"CKVTORSE","created_at":"2026-07-05T06:17:59.048717+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/CKVTORSE5P2MLK5JYKB3KSNRKD","json":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD.json","graph_json":"https://pith.science/api/pith-number/CKVTORSE5P2MLK5JYKB3KSNRKD/graph.json","events_json":"https://pith.science/api/pith-number/CKVTORSE5P2MLK5JYKB3KSNRKD/events.json","paper":"https://pith.science/paper/CKVTORSE"},"agent_actions":{"view_html":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD","download_json":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD.json","view_paper":"https://pith.science/paper/CKVTORSE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.03521&json=true","fetch_graph":"https://pith.science/api/pith-number/CKVTORSE5P2MLK5JYKB3KSNRKD/graph.json","fetch_events":"https://pith.science/api/pith-number/CKVTORSE5P2MLK5JYKB3KSNRKD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD/action/storage_attestation","attest_author":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD/action/author_attestation","sign_citation":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD/action/citation_signature","submit_replication":"https://pith.science/pith/CKVTORSE5P2MLK5JYKB3KSNRKD/action/replication_record"}},"created_at":"2026-07-05T06:17:59.048717+00:00","updated_at":"2026-07-05T06:17:59.048717+00:00"}