{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:35ROWJ2FUS7NZE2BDDPFXEDZR2","short_pith_number":"pith:35ROWJ2F","schema_version":"1.0","canonical_sha256":"df62eb2745a4bedc934118de5b90798ebeb331b718725d61fb2a08bd794e8014","source":{"kind":"arxiv","id":"2005.13420","version":2},"attestation_state":"computed","paper":{"title":"Discretize-Optimize vs. Optimize-Discretize for Time-Series Regression and Continuous Normalizing Flows","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Derek Onken, Lars Ruthotto","submitted_at":"2020-05-27T15:28:11Z","abstract_excerpt":"We compare the discretize-optimize (Disc-Opt) and optimize-discretize (Opt-Disc) approaches for time-series regression and continuous normalizing flows (CNFs) using neural ODEs. Neural ODEs are ordinary differential equations (ODEs) with neural network components. Training a neural ODE is an optimal control problem where the weights are the controls and the hidden features are the states. Every training iteration involves solving an ODE forward and another backward in time, which can require large amounts of computation, time, and memory. Comparing the Opt-Disc and Disc-Opt approaches in image"},"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":"2005.13420","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2020-05-27T15:28:11Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"f1e1188a4d1a77f0554f8313a28a3403662972c358d936c8649c0a04d0830994","abstract_canon_sha256":"b637000f342ae2ac324bc098a208e9005ee4fa1fa1204e1ad67d7ffd545888b7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:23:36.143193Z","signature_b64":"f/IVDOhXpNEsB0yG275Nmcb8EFTnIYsdcbV8yAdS2Q1YXCROhhG8spCWXkJd2BZxLRs+39donT3MgdeQLbcyDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df62eb2745a4bedc934118de5b90798ebeb331b718725d61fb2a08bd794e8014","last_reissued_at":"2026-07-05T01:23:36.142735Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:23:36.142735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Discretize-Optimize vs. Optimize-Discretize for Time-Series Regression and Continuous Normalizing Flows","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Derek Onken, Lars Ruthotto","submitted_at":"2020-05-27T15:28:11Z","abstract_excerpt":"We compare the discretize-optimize (Disc-Opt) and optimize-discretize (Opt-Disc) approaches for time-series regression and continuous normalizing flows (CNFs) using neural ODEs. Neural ODEs are ordinary differential equations (ODEs) with neural network components. Training a neural ODE is an optimal control problem where the weights are the controls and the hidden features are the states. Every training iteration involves solving an ODE forward and another backward in time, which can require large amounts of computation, time, and memory. Comparing the Opt-Disc and Disc-Opt approaches in image"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2005.13420","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2005.13420/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":"2005.13420","created_at":"2026-07-05T01:23:36.142790+00:00"},{"alias_kind":"arxiv_version","alias_value":"2005.13420v2","created_at":"2026-07-05T01:23:36.142790+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2005.13420","created_at":"2026-07-05T01:23:36.142790+00:00"},{"alias_kind":"pith_short_12","alias_value":"35ROWJ2FUS7N","created_at":"2026-07-05T01:23:36.142790+00:00"},{"alias_kind":"pith_short_16","alias_value":"35ROWJ2FUS7NZE2B","created_at":"2026-07-05T01:23:36.142790+00:00"},{"alias_kind":"pith_short_8","alias_value":"35ROWJ2F","created_at":"2026-07-05T01:23:36.142790+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26662","citing_title":"Zero-Shot Size Transfer for Neural ODEs on Sparse Random Graphs: Graphon Limits and Adjoint Convergence","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2502.12396","citing_title":"Scientific Machine Learning of Flow Resistance Using Universal Shallow Water Equations with Differentiable Programming","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2001.04385","citing_title":"Universal Differential Equations for Scientific Machine Learning","ref_index":53,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2","json":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2.json","graph_json":"https://pith.science/api/pith-number/35ROWJ2FUS7NZE2BDDPFXEDZR2/graph.json","events_json":"https://pith.science/api/pith-number/35ROWJ2FUS7NZE2BDDPFXEDZR2/events.json","paper":"https://pith.science/paper/35ROWJ2F"},"agent_actions":{"view_html":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2","download_json":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2.json","view_paper":"https://pith.science/paper/35ROWJ2F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2005.13420&json=true","fetch_graph":"https://pith.science/api/pith-number/35ROWJ2FUS7NZE2BDDPFXEDZR2/graph.json","fetch_events":"https://pith.science/api/pith-number/35ROWJ2FUS7NZE2BDDPFXEDZR2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2/action/storage_attestation","attest_author":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2/action/author_attestation","sign_citation":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2/action/citation_signature","submit_replication":"https://pith.science/pith/35ROWJ2FUS7NZE2BDDPFXEDZR2/action/replication_record"}},"created_at":"2026-07-05T01:23:36.142790+00:00","updated_at":"2026-07-05T01:23:36.142790+00:00"}