{"id":"W3094116343","doi":"10.1007/s10994-023-06337-6","title":"The role of mutual information in variational classifiers","year":2023,"lang":"en","type":"article","venue":"Machine Learning","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"École de Technologie Supérieure","funders":"H2020 Marie Skłodowska-Curie Actions; Universidad de Buenos Aires; Consejo Nacional de Investigaciones Científicas y Técnicas","keywords":"Information bottleneck method; Mutual information; Overfitting; Artificial intelligence; Regularization (linguistics); Computer science; Mathematics; Generalization; Kullback–Leibler divergence; Early stopping; MNIST database; Entropy (arrow of time); Inference; Algorithm; Machine learning; Artificial neural network","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":[0.01132337,0.0009309164,0.002587329,0.001680611,0.001358567,0.002868654,0.004013315,0.003931745,0.001891797],"category_scores_gemma":[0.04829368,0.001685073,0.001351753,0.001334216,0.005508015,0.009407007,0.004487767,0.004549993,0.0002634442],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001706162,"about_ca_system_score_gemma":0.001247751,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003144103,"about_ca_topic_score_gemma":0.002804412,"domain_scores_codex":[0.9949592,0.00318587,0.0001894861,0.0008537255,0.0006167695,0.0001948979],"domain_scores_gemma":[0.9535739,0.04127701,0.001203384,0.002029136,0.00128993,0.000626591],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0001896321,0.0001138306,0.001912188,0.000279824,0.0002620274,0.0001064747,0.0003283166,0.2634072,0.00151139,0.6791852,0.002379453,0.05032434],"study_design_scores_gemma":[0.00001094062,0.00002817006,0.0003181653,0.00002585536,0.00002003104,0.00003868367,0.00001831785,0.6813104,0.0003188479,0.3173572,0.0005271211,0.0000262062],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01850482,0.001751017,0.9760801,0.00150054,0.00009501188,0.00003125451,0.00006556929,0.0001026945,0.001868882],"genre_scores_gemma":[0.7888543,0.003194144,0.1975719,0.001097155,0.001084078,0.0002650188,0.0003766396,0.0004175255,0.007139189],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01132337,"threshold_uncertainty_score":0.05988449,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00828382254537237,"score_gpt":0.2255312153029866,"score_spread":0.2172473927576142,"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."}}