{"id":"W4292707814","doi":"10.3389/fncom.2022.980613","title":"Combining backpropagation with Equilibrium Propagation to improve an Actor-Critic reinforcement learning framework","year":2022,"lang":"en","type":"article","venue":"Frontiers in Computational Neuroscience","topic":"Neural dynamics and brain function","field":"Neuroscience","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mount Royal University; University of Lethbridge","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institutes of Health Research; Compute Canada","keywords":"Backpropagation; Reinforcement learning; Computer science; Artificial intelligence; Variety (cybernetics); Artificial neural network; Task (project management); Machine learning; Propagation of uncertainty; Algorithm; Engineering","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0003987409,0.0001829433,0.0001627526,0.0003568025,0.0006871666,0.0001993412,0.0004255778,0.00002839392,0.00001653653],"category_scores_gemma":[0.0005738435,0.0001861197,0.00002975303,0.001461343,0.0001329282,0.0007548195,0.0002235834,0.0005363645,0.000003870873],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002475957,"about_ca_system_score_gemma":0.0001408306,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007985418,"about_ca_topic_score_gemma":7.776059e-7,"domain_scores_codex":[0.9970485,0.0003189452,0.0003246602,0.0008539336,0.001050869,0.0004031026],"domain_scores_gemma":[0.9992536,0.0001794203,0.000166509,0.0001926487,0.00006550936,0.0001423655],"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":[0.0001277401,0.00007808035,0.001836937,0.000009388147,4.608174e-7,0.00001884502,0.0002440697,0.8853487,0.1068451,0.003770022,0.00004891876,0.00167165],"study_design_scores_gemma":[0.0003720771,0.001669341,0.004497655,0.00001967546,0.000003160249,0.00003009629,0.0001936367,0.9836354,0.003591369,0.005380033,0.0003449246,0.0002626948],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5644624,0.000002255998,0.4321242,0.0007803976,0.001825755,0.0005783498,0.000007012884,0.00009525241,0.0001243353],"genre_scores_gemma":[0.9888147,8.649076e-7,0.007915314,0.002831482,0.0000472359,0.0001416017,0.00002424633,0.00002636668,0.0001981761],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4243523,"threshold_uncertainty_score":0.758974,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01871365039396951,"score_gpt":0.2571294174646655,"score_spread":0.238415767070696,"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."}}