{"id":"W2970667219","doi":"","title":"A Geometric Perspective on Optimal Representations for Reinforcement Learning","year":2019,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Insect and Pesticide Research","field":"Agricultural and Biological Sciences","cited_by":34,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; Université de Montréal","funders":"","keywords":"Reinforcement learning; Computer science; Representation (politics); Bellman equation; Perspective (graphical); Value (mathematics); Class (philosophy); Relaxation (psychology); Domain (mathematical analysis); Function (biology); Adversarial system; Artificial intelligence; Space (punctuation); Mathematical optimization; Machine learning; Mathematics; Mathematical analysis","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":[0.000285871,0.00009507186,0.0001180227,0.00007924132,0.0003407335,0.0003802869,0.0001274522,0.00005802877,0.000129547],"category_scores_gemma":[0.0001927637,0.00004067569,0.00005501264,0.0005527489,0.00001476109,0.0008785441,0.00002228313,0.0001448016,0.0001662802],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009439437,"about_ca_system_score_gemma":0.00001878083,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002503013,"about_ca_topic_score_gemma":0.000002816713,"domain_scores_codex":[0.9989903,0.00003620699,0.0002733768,0.0001234636,0.0003355384,0.0002411156],"domain_scores_gemma":[0.9990882,0.0002201613,0.0001723003,0.00003669753,0.0004287364,0.00005385515],"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":[0.0007125355,0.00009767525,0.007247636,0.0007485803,0.00005177793,0.00000186037,0.00568731,0.7785103,0.01312121,0.01597391,0.003030165,0.174817],"study_design_scores_gemma":[0.0009357855,0.002514085,0.01221817,0.0001919301,0.00001582896,0.00003063719,0.02137196,0.9197664,0.001530899,0.00008861462,0.04085122,0.0004844349],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9707956,0.00009511929,0.0008641193,0.0006251336,0.0003051956,0.001300523,0.00001176687,0.0001414502,0.02586114],"genre_scores_gemma":[0.9978952,0.000003605578,0.00002414052,0.0001049083,0.0001497851,0.0001061441,0.0001226263,8.584342e-7,0.001592715],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.1743325,"threshold_uncertainty_score":0.3667116,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0356229744796313,"score_gpt":0.3060931110972628,"score_spread":0.2704701366176315,"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."}}