{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:SYN2ROAN4MTK6IFIG7OXASZZCH","short_pith_number":"pith:SYN2ROAN","schema_version":"1.0","canonical_sha256":"961ba8b80de326af20a837dd704b3911e0a9d5b0d63ff0145709c30b61417a41","source":{"kind":"arxiv","id":"2310.13901","version":1},"attestation_state":"computed","paper":{"title":"Towards Hyperparameter-Agnostic DNN Training via Dynamical System Insights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SY","eess.SY"],"primary_cat":"cs.LG","authors_text":"Aayushya Agarwal, Bruno Sinopoli, Carmel Fiscko, Larry Pileggi, Soummya Kar, Yihan Ruan","submitted_at":"2023-10-21T03:45:13Z","abstract_excerpt":"We present a stochastic first-order optimization method specialized for deep neural networks (DNNs), ECCO-DNN. This method models the optimization variable trajectory as a dynamical system and develops a discretization algorithm that adaptively selects step sizes based on the trajectory's shape. This provides two key insights: designing the dynamical system for fast continuous-time convergence and developing a time-stepping algorithm to adaptively select step sizes based on principles of numerical integration and neural network structure. The result is an optimizer with performance that is ins"},"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":"2310.13901","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-10-21T03:45:13Z","cross_cats_sorted":["cs.SY","eess.SY"],"title_canon_sha256":"02438534c82bd37bb5aeda487140ea7254b910a93e4efcd41033f6233ad9d44f","abstract_canon_sha256":"40345859e75ec20ba203e9d6ea45347b457c492c1d8c40a47df86602472006a9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:03:36.672710Z","signature_b64":"HROl9A5klcnNLYHQbYXfbHrFKr0FpfCfKLIWQOZJ0ggc90P07h9c1MF+pnnqXPM6u4UzmPhk4fToAQ8D8z16Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"961ba8b80de326af20a837dd704b3911e0a9d5b0d63ff0145709c30b61417a41","last_reissued_at":"2026-07-05T07:03:36.672186Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:03:36.672186Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Hyperparameter-Agnostic DNN Training via Dynamical System Insights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SY","eess.SY"],"primary_cat":"cs.LG","authors_text":"Aayushya Agarwal, Bruno Sinopoli, Carmel Fiscko, Larry Pileggi, Soummya Kar, Yihan Ruan","submitted_at":"2023-10-21T03:45:13Z","abstract_excerpt":"We present a stochastic first-order optimization method specialized for deep neural networks (DNNs), ECCO-DNN. This method models the optimization variable trajectory as a dynamical system and develops a discretization algorithm that adaptively selects step sizes based on the trajectory's shape. This provides two key insights: designing the dynamical system for fast continuous-time convergence and developing a time-stepping algorithm to adaptively select step sizes based on principles of numerical integration and neural network structure. The result is an optimizer with performance that is ins"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.13901","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/2310.13901/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":"2310.13901","created_at":"2026-07-05T07:03:36.672246+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.13901v1","created_at":"2026-07-05T07:03:36.672246+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.13901","created_at":"2026-07-05T07:03:36.672246+00:00"},{"alias_kind":"pith_short_12","alias_value":"SYN2ROAN4MTK","created_at":"2026-07-05T07:03:36.672246+00:00"},{"alias_kind":"pith_short_16","alias_value":"SYN2ROAN4MTK6IFI","created_at":"2026-07-05T07:03:36.672246+00:00"},{"alias_kind":"pith_short_8","alias_value":"SYN2ROAN","created_at":"2026-07-05T07:03:36.672246+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.13397","citing_title":"Learning by solving differential equations","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH","json":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH.json","graph_json":"https://pith.science/api/pith-number/SYN2ROAN4MTK6IFIG7OXASZZCH/graph.json","events_json":"https://pith.science/api/pith-number/SYN2ROAN4MTK6IFIG7OXASZZCH/events.json","paper":"https://pith.science/paper/SYN2ROAN"},"agent_actions":{"view_html":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH","download_json":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH.json","view_paper":"https://pith.science/paper/SYN2ROAN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.13901&json=true","fetch_graph":"https://pith.science/api/pith-number/SYN2ROAN4MTK6IFIG7OXASZZCH/graph.json","fetch_events":"https://pith.science/api/pith-number/SYN2ROAN4MTK6IFIG7OXASZZCH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH/action/storage_attestation","attest_author":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH/action/author_attestation","sign_citation":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH/action/citation_signature","submit_replication":"https://pith.science/pith/SYN2ROAN4MTK6IFIG7OXASZZCH/action/replication_record"}},"created_at":"2026-07-05T07:03:36.672246+00:00","updated_at":"2026-07-05T07:03:36.672246+00:00"}