{"id":"W3102502161","doi":"","title":"Adversarial Soft Advantage Fitting: Imitation Learning without Policy Optimization","year":2020,"lang":"en","type":"article","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; McGill University; Université de Montréal","funders":"","keywords":"Discriminator; Adversarial system; Reinforcement learning; Computer science; Generator (circuit theory); Imitation; Artificial intelligence; Machine learning; Optimization problem; Mathematical optimization; Algorithm; Power (physics); Mathematics","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":[0.0007780724,0.0004435858,0.0004415355,0.0004621246,0.000587481,0.0004764584,0.001365851,0.0003007905,0.00003146566],"category_scores_gemma":[0.003376683,0.0004946494,0.0001896737,0.001482605,0.0000807156,0.001699322,0.0007753462,0.001042619,0.00003867796],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004528914,"about_ca_system_score_gemma":0.0004926096,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001249815,"about_ca_topic_score_gemma":0.00004801482,"domain_scores_codex":[0.9965619,0.0004726343,0.0006185678,0.0008697257,0.0006398574,0.0008372529],"domain_scores_gemma":[0.9978639,0.0002787208,0.0005303586,0.0006616871,0.0001874269,0.0004778834],"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":[0.0000762963,0.00004183661,0.01169042,0.00002598205,0.00002859885,0.00003356411,0.001607884,0.9378433,0.002001729,0.02712319,0.0001874049,0.01933978],"study_design_scores_gemma":[0.0008888987,0.0001986925,0.001442303,0.00003209556,0.00002268364,0.00004088133,0.000208575,0.9933626,0.0007253831,0.0005382937,0.002041873,0.0004977408],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002154714,0.0001158764,0.9760776,0.01742105,0.0002094742,0.0005634019,0.000004136543,0.002424264,0.001029444],"genre_scores_gemma":[0.4971021,0.00003246944,0.4991557,0.002791884,0.0005565943,0.00007109905,0.00002752459,0.00006313391,0.0001995256],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.4949474,"threshold_uncertainty_score":0.9997505,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01128101256007434,"score_gpt":0.250604166278861,"score_spread":0.2393231537187866,"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."}}