{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:OHLHIMCXHGXJRRRBFQF3T5K4KF","short_pith_number":"pith:OHLHIMCX","schema_version":"1.0","canonical_sha256":"71d674305739ae98c6212c0bb9f55c516ac6ce208ecf08cbf82dd5004cf18e9e","source":{"kind":"arxiv","id":"2204.04876","version":2},"attestation_state":"computed","paper":{"title":"Lyapunov-Guided Representation of Recurrent Neural Network Performance","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.DS","nlin.CD","stat.ML"],"primary_cat":"cs.LG","authors_text":"Eli Shlizerman, Ryan Vogt, Yang Zheng","submitted_at":"2022-04-11T05:38:38Z","abstract_excerpt":"Recurrent Neural Networks (RNN) are ubiquitous computing systems for sequences and multivariate time series data. While several robust architectures of RNN are known, it is unclear how to relate RNN initialization, architecture, and other hyperparameters with accuracy for a given task. In this work, we propose to treat RNN as dynamical systems and to correlate hyperparameters with accuracy through Lyapunov spectral analysis, a methodology specifically designed for nonlinear dynamical systems. To address the fact that RNN features go beyond the existing Lyapunov spectral analysis, we propose to"},"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":"2204.04876","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-04-11T05:38:38Z","cross_cats_sorted":["math.DS","nlin.CD","stat.ML"],"title_canon_sha256":"73ae32f9eb2e39ba67d973aa1f4eb3099db6aca7e3ae64a6c27580aa190ba52b","abstract_canon_sha256":"650b8a030b591aef7a2cf6846a01ca82dcb4f3c9fca9741d00b355a25946c0e4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:28:07.823306Z","signature_b64":"jV2kS3bI/BNf4Xfegp4uFDrRTtdNXutCZP++suclN0b7EY6MnGFUB+99D4VFhs7dA+f+pHtzXwRGUoTFl8FeDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71d674305739ae98c6212c0bb9f55c516ac6ce208ecf08cbf82dd5004cf18e9e","last_reissued_at":"2026-07-05T07:28:07.822806Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:28:07.822806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lyapunov-Guided Representation of Recurrent Neural Network Performance","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.DS","nlin.CD","stat.ML"],"primary_cat":"cs.LG","authors_text":"Eli Shlizerman, Ryan Vogt, Yang Zheng","submitted_at":"2022-04-11T05:38:38Z","abstract_excerpt":"Recurrent Neural Networks (RNN) are ubiquitous computing systems for sequences and multivariate time series data. While several robust architectures of RNN are known, it is unclear how to relate RNN initialization, architecture, and other hyperparameters with accuracy for a given task. In this work, we propose to treat RNN as dynamical systems and to correlate hyperparameters with accuracy through Lyapunov spectral analysis, a methodology specifically designed for nonlinear dynamical systems. To address the fact that RNN features go beyond the existing Lyapunov spectral analysis, we propose to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.04876","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/2204.04876/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":"2204.04876","created_at":"2026-07-05T07:28:07.822868+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.04876v2","created_at":"2026-07-05T07:28:07.822868+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.04876","created_at":"2026-07-05T07:28:07.822868+00:00"},{"alias_kind":"pith_short_12","alias_value":"OHLHIMCXHGXJ","created_at":"2026-07-05T07:28:07.822868+00:00"},{"alias_kind":"pith_short_16","alias_value":"OHLHIMCXHGXJRRRB","created_at":"2026-07-05T07:28:07.822868+00:00"},{"alias_kind":"pith_short_8","alias_value":"OHLHIMCX","created_at":"2026-07-05T07:28:07.822868+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.07975","citing_title":"Hyperpruning: Efficient Search through Pruned Variants of Recurrent Neural Networks Leveraging Lyapunov Spectrum","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF","json":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF.json","graph_json":"https://pith.science/api/pith-number/OHLHIMCXHGXJRRRBFQF3T5K4KF/graph.json","events_json":"https://pith.science/api/pith-number/OHLHIMCXHGXJRRRBFQF3T5K4KF/events.json","paper":"https://pith.science/paper/OHLHIMCX"},"agent_actions":{"view_html":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF","download_json":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF.json","view_paper":"https://pith.science/paper/OHLHIMCX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.04876&json=true","fetch_graph":"https://pith.science/api/pith-number/OHLHIMCXHGXJRRRBFQF3T5K4KF/graph.json","fetch_events":"https://pith.science/api/pith-number/OHLHIMCXHGXJRRRBFQF3T5K4KF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF/action/storage_attestation","attest_author":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF/action/author_attestation","sign_citation":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF/action/citation_signature","submit_replication":"https://pith.science/pith/OHLHIMCXHGXJRRRBFQF3T5K4KF/action/replication_record"}},"created_at":"2026-07-05T07:28:07.822868+00:00","updated_at":"2026-07-05T07:28:07.822868+00:00"}