Efficient Reinforcement Learning via Initial Pure Exploration

Sudeep Raja Putta, Theja Tulabandhula

In several realistic situations, an interactive learning agent can practice and refine its strategy before going on to be evaluated. For instance, consider a student preparing for…