{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JNFHQWM6WGH335DVFNZIUGMLK5","short_pith_number":"pith:JNFHQWM6","schema_version":"1.0","canonical_sha256":"4b4a78599eb18fbdf4752b728a198b5755bad6ce00c4159a7751089b5b39608a","source":{"kind":"arxiv","id":"2508.11990","version":1},"attestation_state":"computed","paper":{"title":"Universal Learning of Nonlinear Dynamics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anand Brahmbhatt, Elad Hazan, Evan Dogariu","submitted_at":"2025-08-16T09:14:47Z","abstract_excerpt":"We study the fundamental problem of learning a marginally stable unknown nonlinear dynamical system. We describe an algorithm for this problem, based on the technique of spectral filtering, which learns a mapping from past observations to the next based on a spectral representation of the system. Using techniques from online convex optimization, we prove vanishing prediction error for any nonlinear dynamical system that has finitely many marginally stable modes, with rates governed by a novel quantitative control-theoretic notion of learnability. The main technical component of our method is a"},"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":"2508.11990","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-16T09:14:47Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"fe06fbff865168609287f0f65022543c3b5d85bebeae150a0ed4288e9c4deff1","abstract_canon_sha256":"611ad6286a261057ba82cf4a67e9bbd5a8e1e6d9f8e1b43cfbcc58f70054e616"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:58.357074Z","signature_b64":"XFfx0P40G4Qv7B1F2YDu+zjWzuwJytdzoG6SwDl5uTUXXwspsUcrWtBzsAe9dn/cxCPO1qWQvIlna+S4sgTIBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4b4a78599eb18fbdf4752b728a198b5755bad6ce00c4159a7751089b5b39608a","last_reissued_at":"2026-07-05T11:54:58.356594Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:58.356594Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Universal Learning of Nonlinear Dynamics","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Anand Brahmbhatt, Elad Hazan, Evan Dogariu","submitted_at":"2025-08-16T09:14:47Z","abstract_excerpt":"We study the fundamental problem of learning a marginally stable unknown nonlinear dynamical system. We describe an algorithm for this problem, based on the technique of spectral filtering, which learns a mapping from past observations to the next based on a spectral representation of the system. Using techniques from online convex optimization, we prove vanishing prediction error for any nonlinear dynamical system that has finitely many marginally stable modes, with rates governed by a novel quantitative control-theoretic notion of learnability. The main technical component of our method is a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11990","kind":"arxiv","version":1},"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/2508.11990/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":"2508.11990","created_at":"2026-07-05T11:54:58.356655+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.11990v1","created_at":"2026-07-05T11:54:58.356655+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11990","created_at":"2026-07-05T11:54:58.356655+00:00"},{"alias_kind":"pith_short_12","alias_value":"JNFHQWM6WGH3","created_at":"2026-07-05T11:54:58.356655+00:00"},{"alias_kind":"pith_short_16","alias_value":"JNFHQWM6WGH335DV","created_at":"2026-07-05T11:54:58.356655+00:00"},{"alias_kind":"pith_short_8","alias_value":"JNFHQWM6","created_at":"2026-07-05T11:54:58.356655+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.12691","citing_title":"Two-Layer Linear Auto-Regressive Models Estimate Latent States","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08390","citing_title":"The Power of Second Order Methods for Sequence Preconditioning","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08390","citing_title":"The Power of Second Order Methods for Sequence Preconditioning","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5","json":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5.json","graph_json":"https://pith.science/api/pith-number/JNFHQWM6WGH335DVFNZIUGMLK5/graph.json","events_json":"https://pith.science/api/pith-number/JNFHQWM6WGH335DVFNZIUGMLK5/events.json","paper":"https://pith.science/paper/JNFHQWM6"},"agent_actions":{"view_html":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5","download_json":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5.json","view_paper":"https://pith.science/paper/JNFHQWM6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.11990&json=true","fetch_graph":"https://pith.science/api/pith-number/JNFHQWM6WGH335DVFNZIUGMLK5/graph.json","fetch_events":"https://pith.science/api/pith-number/JNFHQWM6WGH335DVFNZIUGMLK5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5/action/storage_attestation","attest_author":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5/action/author_attestation","sign_citation":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5/action/citation_signature","submit_replication":"https://pith.science/pith/JNFHQWM6WGH335DVFNZIUGMLK5/action/replication_record"}},"created_at":"2026-07-05T11:54:58.356655+00:00","updated_at":"2026-07-05T11:54:58.356655+00:00"}