Chapter 16. Black-Box Optimization in RL
In this chapter, we'll again change our perspective on Reinforcement Learning (RL) training and will switch to the so-called black-box optimizations, in particular the evolution strategies and genetic algorithms. These methods are at least a decade old, but recently several research studies were conducted, which showed the applicability of the methods to large-scale RL problems and their competitiveness with the value iteration and Policy Gradient (PG) methods.