{"id":"W4392681023","doi":"10.1007/s40843-023-2800-x","title":"Enhancing interpretability in the exploration of high-energy conversion efficiency in CsSnBr3−xIx configurations using crystal graph convolutional neural networks and adversarial example methods","year":2024,"lang":"en","type":"article","venue":"Science China Materials","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":6,"is_retracted":false,"has_abstract":false,"ca_institutions":"McGill University","funders":"","keywords":"Interpretability; Artificial intelligence; Classifier (UML); Computer science; Machine learning; Convolutional neural network; Feature engineering; Adversarial system; Artificial neural network; Deep learning; 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":[0.01188913,0.0002462834,0.0004170971,0.000470112,0.0003868852,0.0007242366,0.0007782254,0.00009577368,0.0003917331],"category_scores_gemma":[0.0007662226,0.0001837895,0.00004220812,0.001399947,0.001655553,0.002322865,0.0003012826,0.0001630414,0.000002786363],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001506189,"about_ca_system_score_gemma":0.0002853421,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006751412,"about_ca_topic_score_gemma":0.0001815655,"domain_scores_codex":[0.995547,0.001510898,0.0009034027,0.000854034,0.0006423765,0.0005422907],"domain_scores_gemma":[0.998605,0.0005552672,0.0002624902,0.0004038676,0.00009867725,0.00007473282],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00005266364,0.00003547087,0.0001136176,0.00007143795,0.000001034149,0.000004528469,0.002583477,0.02988988,0.9515054,0.0156294,0.000002002961,0.0001111342],"study_design_scores_gemma":[0.0004266467,0.0001511023,0.013516,0.0002416389,0.00001946989,0.00004632209,0.0008494978,0.3479719,0.6241007,0.01231211,0.00001338518,0.0003512374],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7885396,0.00009555608,0.2079817,0.0001362395,0.002850177,0.0002931558,0.00002748214,0.00004419881,0.00003193599],"genre_scores_gemma":[0.9920956,0.00001123026,0.00767498,0.00004082291,0.0001179674,0.0000309084,0.00001298369,0.00001192181,0.000003584004],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3274046,"threshold_uncertainty_score":0.9998627,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02298444109958607,"score_gpt":0.3143053081534152,"score_spread":0.2913208670538291,"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."}}