{"id":"W2406623448","doi":"10.5555/2615731.2616087","title":"Policy optimization by marginal-map probabilistic inference in generative models","year":2014,"lang":"en","type":"article","venue":"Adaptive Agents and Multi-Agents Systems","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Partially observable Markov decision process; Inference; Computer science; Mathematical optimization; Generative model; Bounded function; Benchmark (surveying); Probabilistic logic; Machine learning; Artificial intelligence; Algorithm; Mathematics; Generative grammar; Markov model; Markov chain","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":[0.0004365471,0.0002701982,0.0003154947,0.0001894682,0.0001794698,0.0002503299,0.0004250907,0.00009178725,0.000005002717],"category_scores_gemma":[0.0001400624,0.0002331631,0.00003741258,0.0003124257,0.00005575934,0.0005186242,0.0002187804,0.0001881699,0.00001789557],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00008824976,"about_ca_system_score_gemma":0.00005898947,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001901074,"about_ca_topic_score_gemma":0.00002375159,"domain_scores_codex":[0.9978576,0.0004472027,0.0003695677,0.0006364839,0.0002973771,0.0003918054],"domain_scores_gemma":[0.9990417,0.0001202265,0.0001927187,0.0003509352,0.0001212966,0.0001730989],"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":[0.000005898322,0.0001474871,0.0009283916,0.00004447163,0.00002544766,0.000005727647,0.001490386,0.9716963,0.00002749147,0.02017974,0.0002979081,0.005150808],"study_design_scores_gemma":[0.0008691485,0.0001152213,0.0008838316,0.00009517078,0.000005503269,0.000003147766,0.00006257046,0.9965168,0.000006108496,0.000344099,0.0008332927,0.0002651038],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004409754,0.0002042871,0.9935559,0.000252152,0.0003250711,0.0005507965,0.00001635414,0.00008184145,0.0006038528],"genre_scores_gemma":[0.9679099,0.00006610774,0.02979838,0.000215262,0.0001037709,0.00009281952,0.00002350684,0.00001895733,0.001771283],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9637575,"threshold_uncertainty_score":0.9508114,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03596804359116691,"score_gpt":0.2878306369712302,"score_spread":0.2518625933800633,"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."}}