{"id":"W4382318225","doi":"10.1609/aaai.v37i13.26968","title":"Towards Safe Reinforcement Learning via OOD Dynamics Detection in Autonomous Driving System (Student Abstract)","year":2023,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Anomaly Detection Techniques and Applications","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Thales (Canada)","funders":"","keywords":"Reinforcement learning; Software deployment; Variety (cybernetics); Relevance (law); Computer science; Dynamics (music); Artificial intelligence; Reinforcement; Function (biology); State (computer science); Human–computer interaction; Psychology; Social psychology; 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009684667,0.0005284753,0.0005976523,0.0003246524,0.0002857328,0.0004870809,0.0007603636,0.0005886974,0.0008611345],"category_scores_gemma":[0.002820504,0.0002905445,0.0003118532,0.0001326776,0.0005418458,0.000461132,0.0009241434,0.000884802,0.0002020676],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005282598,"about_ca_system_score_gemma":0.0008794098,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006446991,"about_ca_topic_score_gemma":0.003830855,"domain_scores_codex":[0.9997494,0.00007241168,0.00001151476,0.00006120162,0.00005440998,0.0000510432],"domain_scores_gemma":[0.9989005,0.0005571041,0.0001271795,0.0000779553,0.0002353828,0.0001018152],"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":[0.0001823462,0.0001434848,0.003708572,0.00005039172,0.00004314078,0.0001411473,0.0001201363,0.8588034,0.006962582,0.003291555,0.001061191,0.125492],"study_design_scores_gemma":[0.000003693475,0.00001984221,0.00007459031,0.000001073664,0.000001084265,0.000003676137,0.000002122674,0.9990153,0.0004129613,0.0004075783,0.00005681316,0.000001158348],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.165525,0.000204172,0.8301618,0.0004069267,0.00005790888,0.00007468295,0.00003890018,0.001898952,0.001631735],"genre_scores_gemma":[0.9380149,0.00002608536,0.06085534,0.00007854473,0.00001251841,0.00004047206,0.00003104989,0.00004089644,0.0009002095],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006446991,"threshold_uncertainty_score":0.01281893,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03409419949122906,"score_gpt":0.2884739143939237,"score_spread":0.2543797149026946,"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."}}