{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LI6O7QGSPVRY4KQZ7ATC5UKI2O","short_pith_number":"pith:LI6O7QGS","schema_version":"1.0","canonical_sha256":"5a3cefc0d27d638e2a19f8262ed148d397ba41898546e6ef89405b4db802ce73","source":{"kind":"arxiv","id":"2407.08641","version":3},"attestation_state":"computed","paper":{"title":"How more data can hurt: Instability and regularization in next-generation reservoir computing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","math.DS","nlin.AO"],"primary_cat":"cs.LG","authors_text":"Edmilson Roque dos Santos, Huixin Zhang, Sean P. Cornelius, Yuanzhao Zhang","submitted_at":"2024-07-11T16:22:13Z","abstract_excerpt":"It has been found recently that more data can, counter-intuitively, hurt the performance of deep neural networks. Here, we show that a more extreme version of the phenomenon occurs in data-driven models of dynamical systems. To elucidate the underlying mechanism, we focus on next-generation reservoir computing (NGRC) -- a popular framework for learning dynamics from data. We find that, despite learning a better representation of the flow map with more training data, NGRC can adopt an ill-conditioned ``integrator'' and lose stability. We link this data-induced instability to the auxiliary dimen"},"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":"2407.08641","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-07-11T16:22:13Z","cross_cats_sorted":["cs.NE","math.DS","nlin.AO"],"title_canon_sha256":"11aef20fddff15aa2cc7598ff1251827e911183785e4f867e47e6a06989b4888","abstract_canon_sha256":"56747966345e027e127fb286546657ba6c7ad5aac702ab2d7b488832ae46f39d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:16.322734Z","signature_b64":"GY0hw9zzstK9B2l2pQpvQcDDud0qrgN4BsDMqocOjfjrHlR9PNGbew3SuvF4NR/I33SWYPwy2tB4fyHJKkVMAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a3cefc0d27d638e2a19f8262ed148d397ba41898546e6ef89405b4db802ce73","last_reissued_at":"2026-07-05T11:32:16.322210Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:16.322210Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How more data can hurt: Instability and regularization in next-generation reservoir computing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.NE","math.DS","nlin.AO"],"primary_cat":"cs.LG","authors_text":"Edmilson Roque dos Santos, Huixin Zhang, Sean P. Cornelius, Yuanzhao Zhang","submitted_at":"2024-07-11T16:22:13Z","abstract_excerpt":"It has been found recently that more data can, counter-intuitively, hurt the performance of deep neural networks. Here, we show that a more extreme version of the phenomenon occurs in data-driven models of dynamical systems. To elucidate the underlying mechanism, we focus on next-generation reservoir computing (NGRC) -- a popular framework for learning dynamics from data. We find that, despite learning a better representation of the flow map with more training data, NGRC can adopt an ill-conditioned ``integrator'' and lose stability. We link this data-induced instability to the auxiliary dimen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08641","kind":"arxiv","version":3},"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/2407.08641/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":"2407.08641","created_at":"2026-07-05T11:32:16.322268+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.08641v3","created_at":"2026-07-05T11:32:16.322268+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08641","created_at":"2026-07-05T11:32:16.322268+00:00"},{"alias_kind":"pith_short_12","alias_value":"LI6O7QGSPVRY","created_at":"2026-07-05T11:32:16.322268+00:00"},{"alias_kind":"pith_short_16","alias_value":"LI6O7QGSPVRY4KQZ","created_at":"2026-07-05T11:32:16.322268+00:00"},{"alias_kind":"pith_short_8","alias_value":"LI6O7QGS","created_at":"2026-07-05T11:32:16.322268+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.31438","citing_title":"Flow map learning in nonlinear vector autoregressive models: influence of the feature-library structure on the training error","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2602.18377","citing_title":"Theory and interpretability of Quantum Extreme Learning Machines: a Pauli-transfer matrix approach","ref_index":112,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O","json":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O.json","graph_json":"https://pith.science/api/pith-number/LI6O7QGSPVRY4KQZ7ATC5UKI2O/graph.json","events_json":"https://pith.science/api/pith-number/LI6O7QGSPVRY4KQZ7ATC5UKI2O/events.json","paper":"https://pith.science/paper/LI6O7QGS"},"agent_actions":{"view_html":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O","download_json":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O.json","view_paper":"https://pith.science/paper/LI6O7QGS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.08641&json=true","fetch_graph":"https://pith.science/api/pith-number/LI6O7QGSPVRY4KQZ7ATC5UKI2O/graph.json","fetch_events":"https://pith.science/api/pith-number/LI6O7QGSPVRY4KQZ7ATC5UKI2O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O/action/storage_attestation","attest_author":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O/action/author_attestation","sign_citation":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O/action/citation_signature","submit_replication":"https://pith.science/pith/LI6O7QGSPVRY4KQZ7ATC5UKI2O/action/replication_record"}},"created_at":"2026-07-05T11:32:16.322268+00:00","updated_at":"2026-07-05T11:32:16.322268+00:00"}