Chandrajit Bajaj, UT Austin Reinforcement learning (RL) with continuous state and action spaces is arguably one the most challenging problems within the field of machine learning.  Most current learning methods… Read More »Global Search and Discovery with Differential Policy Optimization