{"id":"W4402686099","doi":"10.2139/ssrn.4962471","title":"Land Cover Convolutional Neural Network Generalization Across Ecozones Via Training Set Supplementation","year":2024,"lang":"en","type":"preprint","venue":"SSRN Electronic Journal","topic":"Remote Sensing in Agriculture","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Winnipeg; University of Manitoba","funders":"","keywords":"Generalization; Convolutional neural network; Land cover; Set (abstract data type); Cover (algebra); Training (meteorology); Training set; Computer science; Geography; Artificial intelligence; Mathematics; Land use; Meteorology; Engineering; Biology; Ecology","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":[0.001450323,0.0008185686,0.0008182728,0.0005868422,0.0004869149,0.0006325904,0.001249498,0.001316542,0.003473497],"category_scores_gemma":[0.005862359,0.0005713378,0.0007567245,0.0007069019,0.0006932859,0.001553795,0.002036832,0.001360604,0.0006587816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006271893,"about_ca_system_score_gemma":0.0008409779,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01076036,"about_ca_topic_score_gemma":0.01734607,"domain_scores_codex":[0.9996786,0.00007915426,0.00002008564,0.0001320132,0.00004388025,0.00004627695],"domain_scores_gemma":[0.9982136,0.0008225839,0.0001066387,0.0005180108,0.0002764184,0.00006272263],"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":[0.0002947298,0.0001773555,0.006993384,0.0001098818,0.000192539,0.0001619782,0.0001328573,0.7551005,0.009400205,0.00818278,0.006859393,0.2123944],"study_design_scores_gemma":[0.00001083115,0.00002342968,0.0006064,0.000008117662,0.00001735996,0.000008429297,0.00001122134,0.993567,0.0009469386,0.00440321,0.0003937321,0.000003304324],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4606104,0.001081266,0.5181493,0.001743732,0.0002206002,0.0001326096,0.001795438,0.003658159,0.01260868],"genre_scores_gemma":[0.946568,0.0001267675,0.04504227,0.0002686131,0.00005834662,0.0000848922,0.001670679,0.0002040338,0.005976486],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01076036,"threshold_uncertainty_score":0.02139544,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01317982123843856,"score_gpt":0.262298149276289,"score_spread":0.2491183280378505,"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."}}