{"id":"W4400612042","doi":"10.48550/arxiv.2407.08641","title":"How more data can hurt: Instability and regularization in next-generation reservoir computing","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Neural Networks and Reservoir Computing","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Alliance de recherche numérique du Canada; National Science Foundation","keywords":"Instability; Regularization (linguistics); Computer science; Physics; Mechanics; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","open_science"],"consensus_categories":[],"category_scores_codex":[0.0006212634,0.0003076802,0.0003229648,0.0002869742,0.0002210012,0.001015106,0.002113666,0.0002759946,8.830349e-7],"category_scores_gemma":[0.00007508918,0.00032733,0.00006280429,0.001009607,0.0001109061,0.0006701514,0.01383947,0.0009569243,0.000001375273],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001802527,"about_ca_system_score_gemma":0.0001978448,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003005131,"about_ca_topic_score_gemma":0.0007877125,"domain_scores_codex":[0.9971114,0.0002756972,0.0002548205,0.001855965,0.0001360195,0.0003660286],"domain_scores_gemma":[0.9975124,0.0001037265,0.0001940063,0.001942302,0.0001166877,0.000130859],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00001197751,0.00006896663,0.005135384,0.0004376479,0.00005297279,0.0004991212,0.0007489505,0.9268906,0.0001960029,0.05772854,0.0003933648,0.00783652],"study_design_scores_gemma":[0.0001868002,0.00001742555,0.001071941,0.0002265452,0.00002050089,0.000005806059,0.00008584068,0.9723791,0.00003778805,0.02552361,0.0001264172,0.0003182532],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7941922,0.0003675076,0.2012106,0.002746425,0.0007617279,0.0003743073,0.00001843146,0.0002021844,0.0001265921],"genre_scores_gemma":[0.9962384,0.0001271906,0.002698052,0.00007255946,0.0002741057,3.220559e-7,0.0001581414,0.00001914928,0.0004120935],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2020462,"threshold_uncertainty_score":0.9999179,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1738128243145753,"score_gpt":0.2219076627189061,"score_spread":0.04809483840433079,"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."}}