{"id":"W4200056893","doi":"10.1109/iros51168.2021.9636479","title":"A Marginal Log-Likelihood Approach for the Estimation of Discount Factors of Multiple Experts in Inverse Reinforcement Learning","year":2021,"lang":"en","type":"article","venue":"2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)","topic":"Decision-Making and Behavioral Economics","field":"Decision Sciences","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Marginal likelihood; Computer science; Latent variable; Machine learning; Probabilistic logic; Hyperparameter; Expectation–maximization algorithm; Markov decision process; Artificial intelligence; Principle of maximum entropy; Reinforcement learning; Latent variable model; Likelihood function; Variable (mathematics); Markov process; Bayesian probability; Mathematical optimization; Mathematics; Maximum likelihood; Estimation theory; Statistics; Algorithm","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":[],"consensus_categories":[],"category_scores_codex":[0.001354486,0.0001951545,0.0004477028,0.0002717614,0.0001004006,0.0002742655,0.0004859558,0.00009644088,0.0002846694],"category_scores_gemma":[0.001088381,0.0001274674,0.0001596462,0.0002050406,0.0001141598,0.0002091252,0.0001156095,0.0001593753,0.000009211632],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001094817,"about_ca_system_score_gemma":0.0001410898,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004215443,"about_ca_topic_score_gemma":0.0001366349,"domain_scores_codex":[0.9971095,0.0001336142,0.001218675,0.0004725137,0.0008714689,0.0001941893],"domain_scores_gemma":[0.9972167,0.001079259,0.0006488007,0.0003373582,0.0006461123,0.00007179028],"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":[0.0001851932,0.0002822721,0.01710157,0.00004062494,0.00007061606,0.00000515742,0.002844428,0.8998067,0.00184185,0.01034821,0.0003845048,0.06708881],"study_design_scores_gemma":[0.0004012689,0.0001436168,0.001147423,0.0002455058,0.00001542035,0.000006965198,0.01637208,0.9771185,0.002443483,0.001252452,0.0006982483,0.0001550618],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7081993,0.0001110159,0.2884252,0.0001524048,0.001362031,0.0004570924,0.00004522226,0.00000603534,0.001241639],"genre_scores_gemma":[0.9979396,0.0001447081,0.0006227278,0.00001982456,0.00005311175,0.00006329501,0.00005830179,0.00001019818,0.00108816],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.2897404,"threshold_uncertainty_score":0.519797,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2132706932633973,"score_gpt":0.3892692312640018,"score_spread":0.1759985380006045,"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."}}