{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:WKMJUTOSTJZWJ2PMCT2J4DBGJ2","short_pith_number":"pith:WKMJUTOS","schema_version":"1.0","canonical_sha256":"b2989a4dd29a7364e9ec14f49e0c264eb600310b9aad4ca462a18b6319f55ec7","source":{"kind":"arxiv","id":"1910.03193","version":3},"attestation_state":"computed","paper":{"title":"DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"DeepONets learn nonlinear operators from small datasets by splitting input encoding from output evaluation points.","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"George Em Karniadakis, Lu Lu, Pengzhan Jin","submitted_at":"2019-10-08T03:21:14Z","abstract_excerpt":"While it is widely known that neural networks are universal approximators of continuous functions, a less known and perhaps more powerful result is that a neural network with a single hidden layer can approximate accurately any nonlinear continuous operator. This universal approximation theorem is suggestive of the potential application of neural networks in learning nonlinear operators from data. However, the theorem guarantees only a small approximation error for a sufficient large network, and does not consider the important optimization and generalization errors. To realize this theorem in"},"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":"1910.03193","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2019-10-08T03:21:14Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"19440d79637affb4adfee66015557e726c4b03aea5f8ad978d4eb496c4f622fe","abstract_canon_sha256":"37938b4c5e8bb120c0c6d320b5025843878835d5f7e23d0849ce8e8e6ceddb96"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:38:53.699430Z","signature_b64":"IwOwtdg56KMGqj3eu1D+WbqgC3TI3wuopol1uXUX51pUbJkZZ8Thp+O2y4VhyNYv+/LZoEU2vDBFKjNZj5cZBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b2989a4dd29a7364e9ec14f49e0c264eb600310b9aad4ca462a18b6319f55ec7","last_reissued_at":"2026-05-17T23:38:53.698823Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:38:53.698823Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepONet: Learning nonlinear operators for identifying differential equations based on the universal approximation theorem of operators","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"DeepONets learn nonlinear operators from small datasets by splitting input encoding from output evaluation points.","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"George Em Karniadakis, Lu Lu, Pengzhan Jin","submitted_at":"2019-10-08T03:21:14Z","abstract_excerpt":"While it is widely known that neural networks are universal approximators of continuous functions, a less known and perhaps more powerful result is that a neural network with a single hidden layer can approximate accurately any nonlinear continuous operator. This universal approximation theorem is suggestive of the potential application of neural networks in learning nonlinear operators from data. However, the theorem guarantees only a small approximation error for a sufficient large network, and does not consider the important optimization and generalization errors. To realize this theorem in"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We propose deep operator networks (DeepONets) to learn operators accurately and efficiently from a relatively small dataset... we observe high-order error convergence in our computational tests, namely polynomial rates (from half order to fourth order) and even exponential convergence with respect to the training dataset size.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The universal approximation theorem guarantees only a small approximation error for a sufficiently large network, and does not consider the important optimization and generalization errors; the paper assumes these practical errors remain controllable with the branch-trunk split and standard training.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DeepONet learns nonlinear operators for differential equations via branch and trunk sub-networks, achieving high-order error convergence on small datasets.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"DeepONets learn nonlinear operators from small datasets by splitting input encoding from output evaluation points.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"40752807f1f16a5c15a8c012a6bb5ab3d9ca4552911126b9db676b6c5ac0d295"},"source":{"id":"1910.03193","kind":"arxiv","version":3},"verdict":{"id":"e9e894c3-73f8-474a-9205-a4ff7caaefcd","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T03:12:06.249764Z","strongest_claim":"We propose deep operator networks (DeepONets) to learn operators accurately and efficiently from a relatively small dataset... we observe high-order error convergence in our computational tests, namely polynomial rates (from half order to fourth order) and even exponential convergence with respect to the training dataset size.","one_line_summary":"DeepONet learns nonlinear operators for differential equations via branch and trunk sub-networks, achieving high-order error convergence on small datasets.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The universal approximation theorem guarantees only a small approximation error for a sufficiently large network, and does not consider the important optimization and generalization errors; the paper assumes these practical errors remain controllable with the branch-trunk split and standard training.","pith_extraction_headline":"DeepONets learn nonlinear operators from small datasets by splitting input encoding from output evaluation points."},"references":{"count":34,"sample":[{"doi":"","year":2008,"title":"L. Bottou and O. Bousquet. The tradeoffs of large scale learning. InAdvances in Neural Information Processing Systems, pages 161–168, 2008","work_id":"47673b81-1737-4520-91bd-724c4f4c44fa","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2016,"title":"S. L. Brunton, J. L. Proctor, and J. N. Kutz. Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proceedings of the National Academy of Sciences, 113(15)","work_id":"f826ad6f-17c0-4483-a568-39693c824383","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1993,"title":"T. Chen and H. Chen. Approximations of continuous functionals by neural networks with application to dynamic systems.IEEE Transactions on Neural Networks, 4(6):910–918, 1993","work_id":"eb629682-fb82-4e90-8ebd-23bbd07e34bf","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1995,"title":"Approximationcapabilitytofunctionsofseveralvariables,nonlinearfunctionals, and operators by radial basis function neural networks","work_id":"e2d5d285-dc7f-47f9-9d8f-670c4052fb2e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1995,"title":"T.ChenandH.Chen. Universalapproximationtononlinearoperatorsbyneuralnetworkswitharbitrary activation functions and its application to dynamical systems.IEEE Transactions on Neural Networks, 6(4):911–91","work_id":"4dd4bc81-7afb-4634-87a9-29c34b79eb47","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":34,"snapshot_sha256":"19532959f9489164d9df220a7fe1142cba9f75c7ed5a1a4b39726aa340bef3b3","internal_anchors":3},"formal_canon":{"evidence_count":2,"snapshot_sha256":"0d13591aa0d3f0a2e4dcec814a020484f4ba69479275d5d3b78db9e4093b2dc6"},"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":"1910.03193","created_at":"2026-05-17T23:38:53.698916+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.03193v3","created_at":"2026-05-17T23:38:53.698916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.03193","created_at":"2026-05-17T23:38:53.698916+00:00"},{"alias_kind":"pith_short_12","alias_value":"WKMJUTOSTJZW","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_16","alias_value":"WKMJUTOSTJZWJ2PM","created_at":"2026-05-18T12:33:30.264802+00:00"},{"alias_kind":"pith_short_8","alias_value":"WKMJUTOS","created_at":"2026-05-18T12:33:30.264802+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":43,"internal_anchor_count":43,"sample":[{"citing_arxiv_id":"2305.14703","citing_title":"Generative diffusion learning for parametric partial differential equations","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2409.00841","citing_title":"Universal Approximation of Operators with Transformers and Neural Integral Operators","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2503.14568","citing_title":"Teaching Artificial Intelligence to Perform Rapid, Resolution-Invariant Grain Growth Modeling via Fourier Neural Operator","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2505.13510","citing_title":"On the definition and importance of interpretability in scientific machine learning","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22338","citing_title":"Physics-Informed Generative Solver: Bridging Data-Driven Priors and Conservation Laws for Stable Spatiotemporal Field Reconstruction","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22663","citing_title":"Therm-FM: Foundation Model is ALL YOU NEED for 3D-ICs Thermal Simulation","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16594","citing_title":"fPINN-DeepONet: A Physics-Informed Operator Learning Framework for Multi-term Time-fractional Mixed Diffusion-wave Equations","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15881","citing_title":"Symplectic Neural Operators for Learning Infinite Dimensional Hamiltonian Systems","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15285","citing_title":"Universal Approximation of Nonlinear Operators and Their Derivatives","ref_index":83,"is_internal_anchor":true},{"citing_arxiv_id":"2505.18288","citing_title":"Operator Learning for Schr\\\"{o}dinger Equation: Unitarity, Error Bounds, and Time Generalization","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2507.10647","citing_title":"Latent Space Dynamics Identification for Interface Tracking with Application to Shock-Induced Pore Collapse","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2509.23249","citing_title":"Deep Learning for Subspace Regression","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2510.00233","citing_title":"Differentiable Autoencoding Neural Operator for Interpretable and Integrable Latent Space Modeling","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2510.21804","citing_title":"XRePIT: A deep learning-computational fluid dynamics hybrid framework implemented in OpenFOAM for fast, robust, and scalable unsteady simulations","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2510.25781","citing_title":"A Practitioner's Guide to Kolmogorov-Arnold Networks","ref_index":156,"is_internal_anchor":true},{"citing_arxiv_id":"2512.03476","citing_title":"ATHENA: Agentic Team for Hierarchical Evolutionary Numerical Algorithms","ref_index":76,"is_internal_anchor":true},{"citing_arxiv_id":"2512.10723","citing_title":"Generalized Spherical Neural Operators: Green's Function Formulation","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2601.07384","citing_title":"CompNO: A Novel Foundation Model approach for solving Partial Differential Equations","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2602.08318","citing_title":"Is Flow Matching Just Trajectory Replay for Sequential Data?","ref_index":76,"is_internal_anchor":true},{"citing_arxiv_id":"2603.06431","citing_title":"Certified and accurate computation of function space norms of deep neural networks","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04474","citing_title":"Geometry-Aware Neural Optimizer for Shape Optimization and Inversion","ref_index":10,"is_internal_anchor":true},{"citing_arxiv_id":"2604.03289","citing_title":"Toward Artificial Intelligence Enabled Earth System Coupling","ref_index":64,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08935","citing_title":"PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12754","citing_title":"Constraint-Aware Flow Matching: Decision Aligned End-to-End Training for Constrained Sampling","ref_index":59,"is_internal_anchor":true},{"citing_arxiv_id":"2605.11691","citing_title":"Compositional Neural Operators for Multi-Dimensional Fluid Dynamics","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2","json":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2.json","graph_json":"https://pith.science/api/pith-number/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/graph.json","events_json":"https://pith.science/api/pith-number/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/events.json","paper":"https://pith.science/paper/WKMJUTOS"},"agent_actions":{"view_html":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2","download_json":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2.json","view_paper":"https://pith.science/paper/WKMJUTOS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.03193&json=true","fetch_graph":"https://pith.science/api/pith-number/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/graph.json","fetch_events":"https://pith.science/api/pith-number/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/action/storage_attestation","attest_author":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/action/author_attestation","sign_citation":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/action/citation_signature","submit_replication":"https://pith.science/pith/WKMJUTOSTJZWJ2PMCT2J4DBGJ2/action/replication_record"}},"created_at":"2026-05-17T23:38:53.698916+00:00","updated_at":"2026-05-17T23:38:53.698916+00:00"}