{"id":"W2989640809","doi":"10.1609/aaai.v34i04.5857","title":"Algorithmic Improvements for Deep Reinforcement Learning Applied to Interactive Fiction","year":2020,"lang":"en","type":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Artificial Intelligence in Games","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Compute Canada; Canadian Institute for Advanced Research","keywords":"Reinforcement learning; Observability; Computer science; Artificial intelligence; Action (physics); Function (biology); Domain (mathematical analysis); Baseline (sea); Point (geometry); Machine learning; Mathematics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.002570449,0.001166819,0.0009732336,0.0005380926,0.0006147472,0.001089102,0.002198122,0.001530147,0.006219677],"category_scores_gemma":[0.009751167,0.0005399997,0.0006362337,0.0004134769,0.001372956,0.001612658,0.002559236,0.002979898,0.0008392288],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001909931,"about_ca_system_score_gemma":0.002390904,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005443215,"about_ca_topic_score_gemma":0.006960568,"domain_scores_codex":[0.9988382,0.0004127648,0.00006722549,0.0002424448,0.00030962,0.0001296788],"domain_scores_gemma":[0.9970818,0.001869433,0.0001890033,0.0003476883,0.0003535595,0.0001584031],"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":[0.0001401563,0.0002325149,0.001048976,0.0001270736,0.00004420944,0.00006518346,0.0001093872,0.8116859,0.002294343,0.04914522,0.001997116,0.13311],"study_design_scores_gemma":[0.00002076081,0.0000370333,0.00004664594,0.000006809697,0.000004254106,0.00000793398,0.000005155713,0.9884821,0.0003189949,0.01048748,0.0005796918,0.000003185298],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01351759,0.0002172994,0.9803123,0.000419726,0.00006925251,0.0001245476,0.00003600909,0.0007465945,0.004556591],"genre_scores_gemma":[0.5691786,0.0002481538,0.4246893,0.0002930435,0.0001015488,0.0004541119,0.0001453745,0.0001831793,0.004706678],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006219677,"threshold_uncertainty_score":0.02080691,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0794015375588785,"score_gpt":0.3226425396400927,"score_spread":0.2432410020812142,"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."}}