{"id":"W4386566906","doi":"10.18653/v1/2023.eacl-main.50","title":"Self-imitation Learning for Action Generation in Text-based Games","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Atomic Energy of Canada Limited","keywords":"Leverage (statistics); Computer science; Reinforcement learning; Artificial intelligence; Exploit; Pruning; Action (physics); Machine learning; Imitation; Set (abstract data type); Rank (graph theory); Mathematics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002725317,0.00004018408,0.00004174888,0.0001414338,0.00005560246,0.00006696909,0.00009118814,0.00002981232,0.000003208816],"category_scores_gemma":[0.00005645841,0.00004035909,0.00001829806,0.0002704806,0.000001802862,0.0002771468,0.00001594232,0.00004343723,0.00002486121],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004205162,"about_ca_system_score_gemma":0.00003148878,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001532645,"about_ca_topic_score_gemma":0.00006419791,"domain_scores_codex":[0.999492,0.00003166353,0.0001043052,0.0001755871,0.00008968326,0.0001067596],"domain_scores_gemma":[0.9997535,0.00007325668,0.00002798037,0.00009812907,0.00003292984,0.00001425528],"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":[0.000002324233,0.00001837147,0.001029599,0.00002902922,0.000002854256,8.411723e-7,0.0008201648,0.7556776,0.007994651,0.01911516,0.0007114424,0.214598],"study_design_scores_gemma":[0.0001887983,0.00002044518,0.001050523,0.000002523005,7.512405e-7,1.705027e-7,0.0000393223,0.9922156,0.004181439,0.0004683884,0.001782721,0.00004933354],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1093267,0.000003703279,0.8887303,0.001041877,0.000148969,0.0001208442,4.520152e-8,0.0003604614,0.0002670608],"genre_scores_gemma":[0.8447269,0.000002974614,0.1545888,0.0001054146,0.00006733735,0.00004431244,0.00001041712,0.00000383454,0.0004499685],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7354002,"threshold_uncertainty_score":0.1645796,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0835096085731733,"score_gpt":0.3065685559575082,"score_spread":0.2230589473843349,"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."}}