{"id":"W3037179286","doi":"10.65109/rtpw2660","title":"Improving Performance in Reinforcement Learning by Breaking Generalization in Neural Networks","year":2020,"lang":"en","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Reinforcement learning; Computer science; Artificial intelligence; Generalization; Scalability; Artificial neural network; Machine learning","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":[0.000262954,0.0001449428,0.0001473781,0.00009814017,0.00007589744,0.0001742403,0.0004925485,0.00006162991,0.00002013224],"category_scores_gemma":[0.00006183114,0.000149097,0.00002051733,0.0007094463,0.00001311448,0.0008498296,0.0003180279,0.0003725326,0.000008749965],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009697538,"about_ca_system_score_gemma":0.00002151695,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001161978,"about_ca_topic_score_gemma":0.000008129242,"domain_scores_codex":[0.9985412,0.00006526783,0.0004167716,0.0003265101,0.0002540284,0.0003962473],"domain_scores_gemma":[0.9995542,0.00003357547,0.0001406182,0.0001729979,0.00002594138,0.00007269641],"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":[0.000003943954,0.000003034452,0.03566938,0.00001548988,0.000001417746,0.000003438595,0.0004914592,0.9523242,0.0002608611,0.000520713,0.00004758686,0.01065852],"study_design_scores_gemma":[0.000380953,0.0001287072,0.001832135,0.00001832641,9.349348e-7,0.000001409286,0.00002455779,0.9969071,0.0001772597,6.900903e-7,0.0003531919,0.0001747068],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02599528,0.00004364802,0.9717311,0.0004391352,0.0001092073,0.0001613631,1.4208e-8,0.0001476361,0.00137263],"genre_scores_gemma":[0.9939251,0.00002648884,0.004533129,0.001187157,0.0000464934,0.000008915561,0.000009632074,0.00001206683,0.0002510501],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9679298,"threshold_uncertainty_score":0.6080001,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01277579907890778,"score_gpt":0.2148214111520931,"score_spread":0.2020456120731853,"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."}}