{"id":"W4411865843","doi":"10.1063/5.0262977","title":"How more data can hurt: Instability and regularization in next-generation reservoir computing","year":2025,"lang":"en","type":"article","venue":"Chaos An Interdisciplinary Journal of Nonlinear Science","topic":"Neural Networks and Reservoir Computing","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Metropolitan University","funders":"National Institute on Alcohol Abuse and Alcoholism; National Science Foundation","keywords":"Regularization (linguistics); Computer science; Instability; Dynamical systems theory; Integrator; Data-driven; Artificial intelligence; Physics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.00257434,0.0001624158,0.0002712214,0.000529132,0.000691348,0.001140798,0.003053176,0.00006066238,6.070665e-7],"category_scores_gemma":[0.0002144079,0.0001299639,0.00003910842,0.001709196,0.0004087137,0.00396049,0.005317986,0.0004201139,1.350851e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001147907,"about_ca_system_score_gemma":0.0004187272,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000008444479,"about_ca_topic_score_gemma":0.00009728864,"domain_scores_codex":[0.9978288,0.000150088,0.000569759,0.0006286944,0.000480322,0.0003422962],"domain_scores_gemma":[0.9980296,0.00009203405,0.000355091,0.0009354842,0.0004246875,0.0001630826],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002286347,0.001630197,0.03582101,0.000435406,0.00007914598,0.0006817083,0.02619066,0.1196572,0.1012298,0.0146842,0.0009505259,0.6984115],"study_design_scores_gemma":[0.0002884744,0.0001679183,0.007876108,0.0002574406,0.000004124199,0.00009911499,0.0007385043,0.9876922,0.0008498559,0.001832995,0.0000641153,0.0001290901],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8384064,0.0002451616,0.1474078,0.01295203,0.0007844697,0.0001325709,0.000002846184,0.00001957809,0.00004909741],"genre_scores_gemma":[0.9528653,0.00002467261,0.04659535,0.0001312273,0.0003261305,4.129381e-7,0.000006658632,0.00000551259,0.00004471106],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.868035,"threshold_uncertainty_score":0.9998961,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07024895502497869,"score_gpt":0.349315181508553,"score_spread":0.2790662264835743,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}