{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001578725,0.001255295,0.001433523,0.0006308874,0.0004283342,0.0008973073,0.001938716,0.001376128,0.00283585],"category_scores_gemma":[0.009803992,0.000528028,0.0006272114,0.0003687172,0.001310186,0.001954129,0.001148851,0.001845771,0.0005350839],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00112094,"about_ca_system_score_gemma":0.0009697095,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005054253,"about_ca_topic_score_gemma":0.004738271,"domain_scores_codex":[0.9992244,0.0003027692,0.00004602351,0.0002072726,0.0001257917,0.00009375078],"domain_scores_gemma":[0.9946414,0.004132766,0.0004132554,0.0002101972,0.0003543408,0.0002480749],"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":[0.0001222276,0.0001253383,0.00148761,0.0001051048,0.00005500411,0.0001329824,0.0001711806,0.9427096,0.002215457,0.01256204,0.0007836643,0.03952977],"study_design_scores_gemma":[0.000007428196,0.00002348694,0.00005758518,0.00000343155,0.000003457013,0.000008260943,0.000004606403,0.9960321,0.0002041755,0.003546919,0.0001048406,0.000003700785],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08456201,0.0006121595,0.9096997,0.0005603248,0.00008313297,0.0001410043,0.0001061179,0.0008845413,0.003350893],"genre_scores_gemma":[0.9414137,0.0001448491,0.05447868,0.000211649,0.00004444811,0.0001881073,0.0001422558,0.0001105783,0.003265789],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005054253,"threshold_uncertainty_score":0.0100497,"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."}}