{"id":"W4393161074","doi":"10.1609/aaai.v38i19.30101","title":"Solving Non-rectangular Reward-Robust MDPs via Frequency Regularization","year":2024,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Advanced Adaptive Filtering Techniques","field":"Engineering","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Regularization (linguistics); Computer science; Mathematical optimization; Mathematics; Applied mathematics; Artificial intelligence","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":[0.0002831557,0.0002832006,0.0002447386,0.0001919506,0.0001104488,0.0001707519,0.0006859265,0.0001364787,0.000095405],"category_scores_gemma":[0.000178089,0.0002400641,0.0001175968,0.0006555928,0.000179952,0.000437573,0.0001226481,0.0004219569,0.0000735083],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001403424,"about_ca_system_score_gemma":0.0000343285,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001554873,"about_ca_topic_score_gemma":0.000004492536,"domain_scores_codex":[0.9984525,0.000006677027,0.0004846333,0.0003884186,0.0003304823,0.0003373076],"domain_scores_gemma":[0.9992695,0.00004822141,0.00009168377,0.0002318453,0.0002953241,0.00006343973],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00001338086,0.00002469922,0.00003192912,0.0002762008,0.00003387415,0.00000222728,0.0007630712,0.005168069,0.6567591,0.3064633,0.0002511882,0.03021294],"study_design_scores_gemma":[0.000005537438,0.00005457808,0.00002579655,0.0007147272,0.00001466883,0.000005077431,0.00008430283,0.1896433,0.5482986,0.2608646,0.00008795172,0.0002007953],"study_design_candidate":"bench_or_experimental","study_design_consensus":"bench_or_experimental","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05922027,0.0002469712,0.9102672,0.0005435068,0.001095798,0.0007407153,0.00001817947,0.001783269,0.02608405],"genre_scores_gemma":[0.9784774,0.00007457627,0.02096223,0.00002379479,0.0001178677,0.00004949761,0.000001928331,0.00006585494,0.0002268125],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9192572,"threshold_uncertainty_score":0.9789529,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04416407608534164,"score_gpt":0.2634541752053656,"score_spread":0.2192900991200239,"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."}}