Variance reduction for policy gradient with action-dependent factorized baselines / Artificial Intelligence, Research / By hi@aiweekly.co.in