{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:4INHS5MXU4CTF2NA44M3Q76XZI","short_pith_number":"pith:4INHS5MX","schema_version":"1.0","canonical_sha256":"e21a797597a70532e9a0e719b87fd7ca2036c66123e7de14314c9cf8640ccbdf","source":{"kind":"arxiv","id":"1806.07366","version":5},"attestation_state":"computed","paper":{"title":"Neural Ordinary Differential Equations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Deep neural networks can replace discrete layers with continuous dynamics defined by ordinary differential equations.","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Duvenaud, Jesse Bettencourt, Ricky T. Q. Chen, Yulia Rubanova","submitted_at":"2018-06-19T17:50:12Z","abstract_excerpt":"We introduce a new family of deep neural network models. Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a black-box differential equation solver. These continuous-depth models have constant memory cost, adapt their evaluation strategy to each input, and can explicitly trade numerical precision for speed. We demonstrate these properties in continuous-depth residual networks and continuous-time latent variable models. We also construct continuous normalizing flows, "},"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":true,"formal_links_present":true},"canonical_record":{"source":{"id":"1806.07366","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-19T17:50:12Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"cae7d85b5a1ddff55631b70fa27eb07d01a0ad8011cd62c9a6516cbfeae065f8","abstract_canon_sha256":"5c43859e54c56b280e576a8f0d64ee2e5f1514f538f7c5ed10ed70eff5f47a35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:52.504629Z","signature_b64":"gMZ+ikrt+GY2hpe5ha0Od/t7R/2kQk1HcY9yoWaVYPN0S3mo/7XgJXabdYnXSs0XEBu9ffmE8IYhJ1NmlshoCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e21a797597a70532e9a0e719b87fd7ca2036c66123e7de14314c9cf8640ccbdf","last_reissued_at":"2026-05-17T23:38:52.504180Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:52.504180Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Ordinary Differential Equations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Deep neural networks can replace discrete layers with continuous dynamics defined by ordinary differential equations.","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.LG","authors_text":"David Duvenaud, Jesse Bettencourt, Ricky T. Q. Chen, Yulia Rubanova","submitted_at":"2018-06-19T17:50:12Z","abstract_excerpt":"We introduce a new family of deep neural network models. Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a black-box differential equation solver. These continuous-depth models have constant memory cost, adapt their evaluation strategy to each input, and can explicitly trade numerical precision for speed. We demonstrate these properties in continuous-depth residual networks and continuous-time latent variable models. We also construct continuous normalizing flows, "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We introduce a new family of deep neural network models. Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a black-box differential equation solver.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a neural network can be trained to produce a vector field whose integral yields useful representations, and that standard ODE solvers remain numerically stable and differentiable enough for end-to-end gradient descent across the range of problems considered.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Neural networks are redefined as continuous dynamical systems by learning the derivative of the hidden state with a neural network and integrating it with an ODE solver.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Deep neural networks can replace discrete layers with continuous dynamics defined by ordinary differential equations.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"abc96746c8234081dcaad4738c14dd9557ecf382d7526a05e789a62840b98134"},"source":{"id":"1806.07366","kind":"arxiv","version":5},"verdict":{"id":"6da4881e-7ef9-4968-839f-2b0efd2c0cfd","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T12:56:04.870070Z","strongest_claim":"We introduce a new family of deep neural network models. Instead of specifying a discrete sequence of hidden layers, we parameterize the derivative of the hidden state using a neural network. The output of the network is computed using a black-box differential equation solver.","one_line_summary":"Neural networks are redefined as continuous dynamical systems by learning the derivative of the hidden state with a neural network and integrating it with an ODE solver.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a neural network can be trained to produce a vector field whose integral yields useful representations, and that standard ODE solvers remain numerically stable and differentiable enough for end-to-end gradient descent across the range of problems considered.","pith_extraction_headline":"Deep neural networks can replace discrete layers with continuous dynamics defined by ordinary differential equations."},"references":{"count":60,"sample":[{"doi":"","year":2011,"title":"Computationally efficient convolved multiple output G aussian processes","work_id":"a79ccadd-e746-419c-8444-fcba767fd3b6","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2017,"title":"OptNet : Differentiable optimization as a layer in neural networks","work_id":"3b7f8d99-7b4a-4bdb-9e76-305ba58e9a5b","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2013,"title":"A general-purpose software framework for dynamic optimization","work_id":"f244a44e-8889-43b2-be19-547832ef321b","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"CasADi -- A software framework for nonlinear optimization and optimal control","work_id":"c5e1095a-3dfa-41f5-a35c-16af5c5a3219","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2018,"title":"Automatic differentiation in machine learning: a survey","work_id":"8abd3335-a598-4a1c-82f1-435c37825259","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":60,"snapshot_sha256":"81e54cc4e9d8e00e01936c37c564cb69b86b298f26e3b1411a1aa0b3d412b5df","internal_anchors":16},"formal_canon":{"evidence_count":3,"snapshot_sha256":"4941d4f57c642bb9ff2b1966790074421c88e13c5b66503d45885f99b68d26c8"},"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":"1806.07366","created_at":"2026-05-17T23:38:52.504255+00:00"},{"alias_kind":"arxiv_version","alias_value":"1806.07366v5","created_at":"2026-05-17T23:38:52.504255+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.07366","created_at":"2026-05-17T23:38:52.504255+00:00"},{"alias_kind":"pith_short_12","alias_value":"4INHS5MXU4CT","created_at":"2026-05-18T12:32:05.422762+00:00"},{"alias_kind":"pith_short_16","alias_value":"4INHS5MXU4CTF2NA","created_at":"2026-05-18T12:32:05.422762+00:00"},{"alias_kind":"pith_short_8","alias_value":"4INHS5MX","created_at":"2026-05-18T12:32:05.422762+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":40,"internal_anchor_count":40,"sample":[{"citing_arxiv_id":"2605.23887","citing_title":"CHRONOS: Temporally-Aware Multi-Agent Coordination for Evolving Data Marketplaces","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2010.02502","citing_title":"Denoising Diffusion Implicit Models","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2504.13228","citing_title":"Neural Mean-Field Games: Extending Mean-Field Game Theory with Neural Stochastic Differential Equations","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2505.13510","citing_title":"On the definition and importance of interpretability in scientific machine learning","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18360","citing_title":"Nested-GPT for variable-multiplicity parton showers: A case study in the resummation of non-global logarithms","ref_index":64,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11125","citing_title":"Language Modeling with Hyperspherical Flows","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16682","citing_title":"Identify Then Project: Contrastive Learning of Latent Dynamics from Partial Observations with Port-Hamiltonian Structure","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18910","citing_title":"A Tutorial on Symbolic Structural Identifiability Analysis of ODE Models in Julia","ref_index":32,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18360","citing_title":"Nested-GPT for variable-multiplicity parton showers: A case study in the resummation of non-global logarithms","ref_index":62,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18118","citing_title":"Functional Whole-Brain Models: A New Framework for Unifying Brain Structure and Cognitive Function","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15463","citing_title":"Layer-wise Derivative Controlled Networks","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15463","citing_title":"Layer-wise Derivative Controlled Networks","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2507.15774","citing_title":"Time Series Forecasting Through the Lens of Dynamics","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2509.04154","citing_title":"Robust Filter Attention: Self-Attention as Precision-Weighted State Estimation","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2509.20430","citing_title":"pop-cosmos: Star formation over 12 Gyr from generative modelling of a deep infrared-selected galaxy catalogue","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2510.06508","citing_title":"Application of deep neural networks for computing the renormalization group flow of the two-dimensional phi^4 field theory","ref_index":50,"is_internal_anchor":true},{"citing_arxiv_id":"2510.25781","citing_title":"A Practitioner's Guide to Kolmogorov-Arnold Networks","ref_index":237,"is_internal_anchor":true},{"citing_arxiv_id":"2511.07686","citing_title":"Kolmogorov-Arnold Chemical Reaction Neural Networks for learning pressure-dependent kinetic rate laws","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2511.16520","citing_title":"Saving Foundation Flow-Matching Priors for Inverse Problems","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2512.12116","citing_title":"Neural CDEs as Correctors for Learned Time Series Models","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2512.15628","citing_title":"Learning continuous state of charge dependent thermal decomposition kinetics for Li-ion cathodes using Kolmogorov-Arnold Chemical Reaction Neural Networks (KA-CRNNs)","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2512.15692","citing_title":"mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2604.04957","citing_title":"FluxMC: Rapid and High-Fidelity Inference for Space-Based Gravitational-Wave Observations","ref_index":59,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03511","citing_title":"Monte Carlo Event Generation with Continuous Normalizing Flows","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11125","citing_title":"Language Modeling with Hyperspherical Flows","ref_index":12,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":3,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI","json":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI.json","graph_json":"https://pith.science/api/pith-number/4INHS5MXU4CTF2NA44M3Q76XZI/graph.json","events_json":"https://pith.science/api/pith-number/4INHS5MXU4CTF2NA44M3Q76XZI/events.json","paper":"https://pith.science/paper/4INHS5MX"},"agent_actions":{"view_html":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI","download_json":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI.json","view_paper":"https://pith.science/paper/4INHS5MX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1806.07366&json=true","fetch_graph":"https://pith.science/api/pith-number/4INHS5MXU4CTF2NA44M3Q76XZI/graph.json","fetch_events":"https://pith.science/api/pith-number/4INHS5MXU4CTF2NA44M3Q76XZI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI/action/storage_attestation","attest_author":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI/action/author_attestation","sign_citation":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI/action/citation_signature","submit_replication":"https://pith.science/pith/4INHS5MXU4CTF2NA44M3Q76XZI/action/replication_record"}},"created_at":"2026-05-17T23:38:52.504255+00:00","updated_at":"2026-05-17T23:38:52.504255+00:00"}