{"id":"W2884439071","doi":"10.1017/s0269888921000035","title":"Safe option-critic: learning safety in the option-critic architecture","year":2021,"lang":"en","type":"article","venue":"The Knowledge Engineering Review","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Reinforcement learning; Consistency (knowledge bases); State space; Variance (accounting); Function (biology); Artificial intelligence; Grid; Architecture; Machine learning; Mathematical optimization; Mathematics","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":[0.001643789,0.000252554,0.0003484565,0.00007756652,0.0002174392,0.0001816222,0.001359793,0.00005676823,0.00004580128],"category_scores_gemma":[0.001389107,0.000167739,0.0001574795,0.00114716,0.00004102598,0.0001792027,0.0003206074,0.0009715615,0.0002983311],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009484505,"about_ca_system_score_gemma":0.0001016009,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000002195289,"about_ca_topic_score_gemma":0.000002727673,"domain_scores_codex":[0.9979494,0.0004507728,0.0004974639,0.0003397594,0.000324367,0.0004382],"domain_scores_gemma":[0.9978253,0.0009706583,0.00006709701,0.0009417705,0.000131885,0.0000632521],"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":[9.844007e-7,0.00003285973,0.00001955698,0.002182542,0.0000271462,0.00003877484,0.001842613,0.8661623,0.0001145144,0.08294491,0.0002565392,0.04637726],"study_design_scores_gemma":[0.0002249683,0.00004013082,0.0007308767,0.005541911,0.00005358993,0.000249198,0.00005991611,0.6134055,0.000042816,0.0002552715,0.3790195,0.000376346],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0001052779,0.1021311,0.8887724,0.00449946,0.0004756741,0.0003700208,4.166853e-7,0.0001599739,0.003485671],"genre_scores_gemma":[0.6906933,0.1747999,0.1154852,0.005543835,0.001801255,0.0005579196,0.00009113982,0.0002463204,0.01078114],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7732872,"threshold_uncertainty_score":0.6840199,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01161079066951347,"score_gpt":0.255187380824089,"score_spread":0.2435765901545756,"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."}}