{"id":"W6891630658","doi":"10.48448/5txp-gh52","title":"TAC: Towered Actor Critic for Handling Multiple Action Types in Reinforcement Learning for Drug Discovery","year":2021,"lang":"en","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta; Polytechnique Montréal","funders":"","keywords":"Reinforcement learning; Action (physics); Drug discovery; Control (management); Action selection; Monotonic function","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":[0.001828647,0.001449526,0.0008613405,0.0006098685,0.0005098575,0.0008681518,0.002041033,0.001312269,0.008226646],"category_scores_gemma":[0.005021726,0.0004362182,0.0007823823,0.0005403628,0.001004713,0.001156607,0.001395497,0.002448767,0.001584152],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001099374,"about_ca_system_score_gemma":0.001959969,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009424592,"about_ca_topic_score_gemma":0.01215661,"domain_scores_codex":[0.9993006,0.0002500164,0.00004392599,0.000158078,0.0001790792,0.00006826644],"domain_scores_gemma":[0.9984408,0.000955447,0.0001254221,0.0001611705,0.0001951691,0.0001219476],"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":[0.0002547414,0.00008564203,0.0006304747,0.0001638017,0.00006952031,0.000146151,0.0000557058,0.8777187,0.001814331,0.01531487,0.009382992,0.09436311],"study_design_scores_gemma":[0.0000229237,0.00002197603,0.00003317386,0.000006805058,0.000006838811,0.0000144541,0.000002295596,0.9939075,0.0005081233,0.004280703,0.001189808,0.000005346611],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009409911,0.0005903222,0.9743946,0.0003907228,0.0002008777,0.0001119455,0.0002723277,0.006211613,0.008417704],"genre_scores_gemma":[0.5387841,0.0005141203,0.445185,0.0005988522,0.0001704661,0.0003709535,0.00100936,0.001034665,0.01233258],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009424592,"threshold_uncertainty_score":0.02752084,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07687368310584854,"score_gpt":0.3634088448580373,"score_spread":0.2865351617521888,"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."}}