{"id":"W3037298096","doi":"10.3389/frai.2020.609673","title":"EmotionNet Nano: An Efficient Deep Convolutional Neural Network Design for Real-Time Facial Expression Recognition","year":2021,"lang":"en","type":"preprint","venue":"Frontiers in Artificial Intelligence","topic":"Emotion and Mood Recognition","field":"Psychology","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Microsoft","keywords":"Convolutional neural network; Deep learning; Computer science; Bottleneck; Software deployment; Benchmark (surveying); Artificial intelligence; Artificial neural network; Facial expression; Computer engineering; Inference; Recurrent neural network; Embedded system; Machine learning; Computer architecture; Real-time computing; Software engineering","routes":{"ca_aff":true,"ca_fund":true,"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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001217694,0.00050151,0.0006408276,0.0003729549,0.0003286926,0.0001892605,0.0004188124,0.0009536138,0.001646126],"category_scores_gemma":[0.0002265921,0.0005798366,0.0003122558,0.0003575427,0.0002266598,0.0001437157,0.0001739395,0.0007314643,0.0001789872],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002821013,"about_ca_system_score_gemma":0.0001785969,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00009847534,"about_ca_topic_score_gemma":0.0000798956,"domain_scores_codex":[0.995111,0.001167546,0.001177896,0.00136599,0.000386565,0.0007909971],"domain_scores_gemma":[0.9980691,0.0002215577,0.0004945901,0.0005370606,0.000456102,0.0002215976],"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":[0.002049317,0.001612592,0.0001637262,0.0001195277,0.0001245452,0.00004332072,0.005069788,0.5767622,0.002651887,0.0006376129,0.0115883,0.3991772],"study_design_scores_gemma":[0.000420255,0.0005332365,0.000785702,0.0008154378,0.0001568255,0.00002522441,0.005501557,0.8939123,0.01103887,0.085146,0.0002405412,0.001424103],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1005169,0.0003756779,0.8768489,0.000112378,0.01925336,0.001911871,0.0001773517,0.0001913138,0.0006122482],"genre_scores_gemma":[0.671719,0.0002341915,0.3147159,0.0003000658,0.003828727,0.001426811,0.007102654,0.0001640716,0.0005086389],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5712021,"threshold_uncertainty_score":0.9996653,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09645069983990581,"score_gpt":0.3314145790208798,"score_spread":0.234963879180974,"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."}}