{"id":"W4362458506","doi":"10.1146/annurev-vision-120522-031739","title":"Are Deep Neural Networks Adequate Behavioral Models of Human Visual Perception?","year":2023,"lang":"en","type":"review","venue":"Annual Review of Vision Science","topic":"Visual perception and processing mechanisms","field":"Neuroscience","cited_by":69,"is_retracted":false,"has_abstract":true,"ca_institutions":"Google (Canada)","funders":"Deutsche Forschungsgemeinschaft","keywords":"Computer science; Artificial intelligence; CLARITY; Perception; Cognitive neuroscience of visual object recognition; Computational model; Object (grammar); Machine learning; Deep neural networks; Artificial neural network; Psychology; Neuroscience","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.002597081,0.0006021078,0.00246828,0.0006361105,0.0004948256,0.0001118957,0.002054367,0.000257415,0.0003955718],"category_scores_gemma":[0.0006581404,0.000456159,0.0008094899,0.004243659,0.001227341,0.001139061,0.0007213036,0.0006183302,0.0001693829],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009709011,"about_ca_system_score_gemma":0.0001991895,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001343589,"about_ca_topic_score_gemma":0.000002392092,"domain_scores_codex":[0.9932888,0.0005994798,0.002041773,0.001368294,0.002019687,0.000681971],"domain_scores_gemma":[0.9953269,0.0002209688,0.002708443,0.0008100743,0.0006119198,0.0003216573],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"systematic_review","study_design_scores_codex":[0.000003006284,0.0001983003,2.8427e-7,0.03542415,0.000002805153,0.00001255317,0.0002221964,0.00002945271,0.0002968863,0.0003975527,0.0002004162,0.9632124],"study_design_scores_gemma":[0.0008000092,0.004902142,0.0001022158,0.8345002,0.002221318,0.0003106847,0.001844361,0.05711389,0.000165891,0.0009605141,0.09270325,0.004375524],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.0002248237,0.995882,0.001331318,0.0000267238,0.0008911539,0.001183651,0.0001043959,0.0001362464,0.0002197062],"genre_scores_gemma":[0.005247985,0.9936717,0.00007629756,0.0004570067,0.0001252929,0.00007187473,0.00002170754,0.00007202731,0.0002561801],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.9588369,"threshold_uncertainty_score":0.999789,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2077491452195851,"score_gpt":0.5099567036521657,"score_spread":0.3022075584325806,"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."}}