{"id":"W2785940258","doi":"","title":"An inference-based policy gradient method for learning options","year":2018,"lang":"en","type":"article","venue":"UvA-DARE (University of Amsterdam)","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Inference; Abstraction; Scalability; Artificial intelligence; Machine learning; A priori and a posteriori; Formalism (music); Differentiable function","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":[0.0003776441,0.0001473401,0.0002121078,0.0003985172,0.0005637829,0.00008333642,0.001161167,0.00008841848,0.00006598506],"category_scores_gemma":[0.0001120574,0.000181768,0.0001264271,0.0005450328,0.0001777144,0.0006405927,0.0002423926,0.0001574878,0.00004168008],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001153118,"about_ca_system_score_gemma":0.0002492683,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003557699,"about_ca_topic_score_gemma":0.00005444857,"domain_scores_codex":[0.9987137,0.0001609712,0.0001445242,0.0003759705,0.0002626546,0.0003421586],"domain_scores_gemma":[0.9984515,0.0001992877,0.0002224922,0.000588349,0.000379775,0.0001586284],"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":[0.0001141697,0.0002052788,0.0042772,0.0001395071,0.0001248319,0.00001346302,0.01537275,0.7170907,0.00407284,0.1705004,0.0009972716,0.08709162],"study_design_scores_gemma":[0.0008045773,0.001244293,0.003954963,0.0000466656,0.00002379842,0.000002202068,0.0007793839,0.9759791,0.0004018248,0.0006120962,0.01591606,0.0002349955],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007715416,0.000004474573,0.9885425,0.001066562,0.0001574581,0.0002237771,0.000005874993,0.0001916445,0.002092281],"genre_scores_gemma":[0.5074362,0.000002552146,0.4912693,0.0001299693,0.00007372131,5.556398e-7,0.00001778381,0.000008923055,0.001060974],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.4997208,"threshold_uncertainty_score":0.7412283,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02574182283153252,"score_gpt":0.2984720200177832,"score_spread":0.2727301971862506,"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."}}