{"id":"W4381194140","doi":"10.1016/j.infsof.2023.107281","title":"Robustness assessment of hyperspectral image CNNs using metamorphic testing","year":2023,"lang":"en","type":"article","venue":"Information and Software Technology","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":false,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Robustness (evolution); Computer science; Convolutional neural network; Artificial intelligence; Hyperspectral imaging; Machine learning; Deep learning; Leverage (statistics); Pattern recognition (psychology)","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":[],"consensus_categories":[],"category_scores_codex":[0.000388253,0.00010964,0.0001893075,0.0007376877,0.000167258,0.00006802646,0.0004359366,0.0001228931,0.000005606043],"category_scores_gemma":[0.0007855959,0.0001077038,0.00002713419,0.00168932,0.0001281546,0.001350341,0.0004474681,0.000252754,0.000004430744],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004384687,"about_ca_system_score_gemma":0.00009855816,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002029594,"about_ca_topic_score_gemma":9.526593e-7,"domain_scores_codex":[0.9990836,0.00002747076,0.0003335769,0.0001374546,0.0001879726,0.0002298898],"domain_scores_gemma":[0.9991172,0.0001386995,0.0002366915,0.0002950611,0.0001804273,0.0000319418],"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":[0.000003890677,0.0000277299,0.06413767,0.0001910463,0.00005748696,0.00002654742,0.0009509728,0.5644845,0.002534949,0.1338529,0.00006610774,0.2336662],"study_design_scores_gemma":[0.0002515083,0.00003782487,0.00571113,0.00002268495,0.000007996932,0.00005570031,0.0002467312,0.9916307,0.0002615565,0.001586574,0.00006558279,0.0001219661],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1882303,0.00001031569,0.8098637,0.0004148417,0.0001187436,0.00008907511,0.000001584355,0.001085952,0.0001854101],"genre_scores_gemma":[0.4036568,0.000003813178,0.5962859,0.00002799011,0.000007876377,0.000005245884,0.000003604323,0.00000410729,0.000004670932],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4271462,"threshold_uncertainty_score":0.4392035,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02552026359889529,"score_gpt":0.2976550759759764,"score_spread":0.2721348123770811,"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."}}