{"id":"W3037476194","doi":"10.1609/icaps.v30i1.6750","title":"Symbolic Plans as High-Level Instructions for Reinforcement Learning","year":2020,"lang":"en","type":"article","venue":"Proceedings of the International Conference on Automated Planning and Scheduling","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":63,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; Agencia Nacional de Investigación y Desarrollo","keywords":"Reinforcement learning; Computer science; Maximization; Task (project management); State (computer science); Action (physics); Artificial intelligence; Machine learning; Mathematical optimization; Programming language; 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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002061071,0.0001560623,0.0001699397,0.0001040279,0.000259546,0.0003147045,0.0009181796,0.00007120496,0.000009760322],"category_scores_gemma":[0.0005754906,0.0001293526,0.00005550437,0.0001791442,0.00004782444,0.0003155472,0.0002764102,0.0003104881,0.000006213363],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003268395,"about_ca_system_score_gemma":0.00006758879,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001223416,"about_ca_topic_score_gemma":9.037417e-8,"domain_scores_codex":[0.9987517,0.000006378486,0.0003285032,0.0002922349,0.0004261646,0.0001950217],"domain_scores_gemma":[0.999072,0.00008251376,0.0003432672,0.00007169991,0.0003472561,0.00008329513],"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":[0.00004665603,0.0000074188,0.002512356,0.00005390745,0.00007931708,4.933848e-7,0.001980237,0.6440079,0.01048321,0.3402543,0.0001620346,0.0004121316],"study_design_scores_gemma":[0.0004229061,0.0001898126,0.0008624433,0.0002901545,0.00001180751,0.00001132982,0.0005378793,0.9882947,0.008106588,0.0008705072,0.000259129,0.0001428082],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.631974,0.0000177262,0.3408806,0.01254832,0.0009165037,0.0004908505,0.000009777928,0.001017352,0.01214487],"genre_scores_gemma":[0.977276,0.000009280958,0.02187823,0.000455935,0.00006897823,0.00001698385,0.000007237454,0.00001079374,0.0002765903],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.345302,"threshold_uncertainty_score":0.5274847,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06793028415246624,"score_gpt":0.2946574244530109,"score_spread":0.2267271403005447,"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."}}