{"id":"W2099759795","doi":"10.1109/tnn.2005.851786","title":"Constructive Feedforward Neural Networks Using Hermite Polynomial Activation Functions","year":2005,"lang":"en","type":"article","venue":"IEEE Transactions on Neural Networks","topic":"Neural Networks and Applications","field":"Computer Science","cited_by":142,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Activation function; Constructive; Sigmoid function; Hermite polynomials; Artificial neural network; Computer science; Feedforward neural network; Orthonormal basis; Polynomial; Adaptation (eye); Function (biology); Constructive proof; Topology (electrical circuits); Mathematics; Artificial intelligence; Discrete mathematics; Pure mathematics; Combinatorics; Mathematical analysis","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00110199,0.0009758775,0.0004992142,0.0004229008,0.0002675553,0.0007228243,0.001509367,0.00111485,0.00140234],"category_scores_gemma":[0.003307554,0.000445635,0.0005185933,0.0004350083,0.0007855488,0.001471989,0.001027719,0.001102258,0.0007661518],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005393652,"about_ca_system_score_gemma":0.0005060744,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0009381393,"about_ca_topic_score_gemma":0.001635289,"domain_scores_codex":[0.9994715,0.0002095742,0.00001999065,0.00007048185,0.0001886838,0.00003975152],"domain_scores_gemma":[0.9989441,0.0006172421,0.00009100555,0.00009020782,0.0002285143,0.00002893404],"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":[0.0001089935,0.00005132554,0.000351568,0.0001375307,0.00005423497,0.0001936839,0.0001139337,0.7953232,0.02000454,0.04060056,0.0009390399,0.1421214],"study_design_scores_gemma":[0.000003859045,0.00002708088,0.00004321602,0.000007512977,0.000007408853,0.00003595216,0.000004325513,0.990364,0.004249271,0.004561508,0.0006888492,0.000006973434],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004922759,0.0001258555,0.9934167,0.00004485226,0.00002308709,0.00001381864,0.000009732321,0.0002517358,0.001191513],"genre_scores_gemma":[0.5174772,0.0005659182,0.4753587,0.0001703323,0.00006871075,0.0001427584,0.00009335914,0.0001059378,0.006017128],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001509367,"threshold_uncertainty_score":0.005827904,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01768520251766669,"score_gpt":0.2410274838492189,"score_spread":0.2233422813315522,"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."}}