{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:J3KBA6XRPNA5KJYWNLI3TCRZXR","short_pith_number":"pith:J3KBA6XR","schema_version":"1.0","canonical_sha256":"4ed4107af17b41d527166ad1b98a39bc6f0e16fbf9f5dd891c5a8347c1f70be0","source":{"kind":"arxiv","id":"1909.13334","version":2},"attestation_state":"computed","paper":{"title":"Symplectic Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jianyu Zhang, L\\'eon Bottou, Martin Arjovsky, Zhengdao Chen","submitted_at":"2019-09-29T18:04:07Z","abstract_excerpt":"We propose Symplectic Recurrent Neural Networks (SRNNs) as learning algorithms that capture the dynamics of physical systems from observed trajectories. An SRNN models the Hamiltonian function of the system by a neural network and furthermore leverages symplectic integration, multiple-step training and initial state optimization to address the challenging numerical issues associated with Hamiltonian systems. We show that SRNNs succeed reliably on complex and noisy Hamiltonian systems. We also show how to augment the SRNN integration scheme in order to handle stiff dynamical systems such as bou"},"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":"1909.13334","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-09-29T18:04:07Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"7e1f76ac3627cd55164aacf079e0b4a27c473ebd65086457635b3f3c195dbf67","abstract_canon_sha256":"4a2323f9335af48e4e6c5972df611c2b231eca68c7f5b8c8c32b6414001614dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:58:05.159976Z","signature_b64":"k9mbIffJKhymBwqEBGnksDo5yVpIY+a7MYXSjkwuZE3ppQv5MV7mOEYYWcWywq/qL7TssXB5T0uZoeyr6tHACQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ed4107af17b41d527166ad1b98a39bc6f0e16fbf9f5dd891c5a8347c1f70be0","last_reissued_at":"2026-07-05T00:58:05.159564Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:58:05.159564Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Symplectic Recurrent Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Jianyu Zhang, L\\'eon Bottou, Martin Arjovsky, Zhengdao Chen","submitted_at":"2019-09-29T18:04:07Z","abstract_excerpt":"We propose Symplectic Recurrent Neural Networks (SRNNs) as learning algorithms that capture the dynamics of physical systems from observed trajectories. An SRNN models the Hamiltonian function of the system by a neural network and furthermore leverages symplectic integration, multiple-step training and initial state optimization to address the challenging numerical issues associated with Hamiltonian systems. We show that SRNNs succeed reliably on complex and noisy Hamiltonian systems. We also show how to augment the SRNN integration scheme in order to handle stiff dynamical systems such as bou"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.13334","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/1909.13334/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":"1909.13334","created_at":"2026-07-05T00:58:05.159622+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.13334v2","created_at":"2026-07-05T00:58:05.159622+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.13334","created_at":"2026-07-05T00:58:05.159622+00:00"},{"alias_kind":"pith_short_12","alias_value":"J3KBA6XRPNA5","created_at":"2026-07-05T00:58:05.159622+00:00"},{"alias_kind":"pith_short_16","alias_value":"J3KBA6XRPNA5KJYW","created_at":"2026-07-05T00:58:05.159622+00:00"},{"alias_kind":"pith_short_8","alias_value":"J3KBA6XR","created_at":"2026-07-05T00:58:05.159622+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27029","citing_title":"Symplectic Neural Networks for learning Generalized Hamiltonians","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18937","citing_title":"Symplecticity-preserving prediction of parameter-dependent Hamiltonian dynamics by Generalized Kernel Interpolation","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23510","citing_title":"Learning partially observed systems with neural Hamiltonian ordinary differential equations","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2507.14491","citing_title":"Artifacts of Numerical Integration in Learning Dynamical Systems","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2603.17493","citing_title":"DustNET: enabling machine learning and AI models of dusty plasmas","ref_index":234,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR","json":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR.json","graph_json":"https://pith.science/api/pith-number/J3KBA6XRPNA5KJYWNLI3TCRZXR/graph.json","events_json":"https://pith.science/api/pith-number/J3KBA6XRPNA5KJYWNLI3TCRZXR/events.json","paper":"https://pith.science/paper/J3KBA6XR"},"agent_actions":{"view_html":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR","download_json":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR.json","view_paper":"https://pith.science/paper/J3KBA6XR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.13334&json=true","fetch_graph":"https://pith.science/api/pith-number/J3KBA6XRPNA5KJYWNLI3TCRZXR/graph.json","fetch_events":"https://pith.science/api/pith-number/J3KBA6XRPNA5KJYWNLI3TCRZXR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR/action/storage_attestation","attest_author":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR/action/author_attestation","sign_citation":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR/action/citation_signature","submit_replication":"https://pith.science/pith/J3KBA6XRPNA5KJYWNLI3TCRZXR/action/replication_record"}},"created_at":"2026-07-05T00:58:05.159622+00:00","updated_at":"2026-07-05T00:58:05.159622+00:00"}