{"id":"W4313483332","doi":"10.48550/arxiv.2301.00051","title":"Learning from Guided Play: Improving Exploration for Adversarial Imitation Learning with Simple Auxiliary Tasks","year":2022,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Toronto Rehabilitation Institute; University of Alberta","funders":"","keywords":"Leverage (statistics); Computer science; Artificial intelligence; Machine learning; Reinforcement learning; Task (project management); Adversarial system; Imitation; Engineering","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":[0.001953252,0.001538965,0.001239527,0.0004166361,0.0003814795,0.000771966,0.002094567,0.00139807,0.002584432],"category_scores_gemma":[0.008356214,0.0006583859,0.0006803624,0.0002945126,0.001899769,0.001854908,0.003002337,0.002518751,0.0006431442],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007755317,"about_ca_system_score_gemma":0.001185201,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002530304,"about_ca_topic_score_gemma":0.002566657,"domain_scores_codex":[0.9992309,0.0003037252,0.00003545005,0.0001674559,0.0001582938,0.0001041399],"domain_scores_gemma":[0.9966254,0.002318746,0.0002689865,0.0004128857,0.0001843157,0.0001897658],"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":[0.0001980578,0.0001477372,0.001490159,0.0001255029,0.0000620369,0.0001282668,0.0001309883,0.9293042,0.003165559,0.01229671,0.001397549,0.05155319],"study_design_scores_gemma":[0.00001165027,0.00005292766,0.00006291667,0.000008006306,0.000004259487,0.00001506371,0.000005458857,0.9940401,0.0003886108,0.005189388,0.0002173685,0.000004184844],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03951138,0.000533326,0.9550472,0.0003264173,0.00005119541,0.00009441185,0.00006856029,0.001376037,0.002991401],"genre_scores_gemma":[0.9046053,0.0002495081,0.09081595,0.0003034071,0.00005670681,0.0002193994,0.0001804145,0.000206042,0.00336321],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002584432,"threshold_uncertainty_score":0.01032996,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09622267453762151,"score_gpt":0.2116297504875747,"score_spread":0.1154070759499532,"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."}}