Reinforcement Learning Using Adaptive Search Method

Abstract

We propose an adaptive probability density function (PDF) to select an effective action on reinforcement learning (RL). The uniform distribution function and the normal distribution function of an action are often used to select an action. When these functions are used, however, the information of search direction is not considered. The proposed method utilizing the information of it enables RL to reduce the number of trials, which is needed to real environment learning. Furthermore, the proposed method can be applied easily to various methods of RL, for example, actor-critic, stochastic gradient ascent method. The performance of our proposed method is demonstrated by computer simulations.

Evolutionary reinforcement learning system with time-varying parameters

“…When using such a selection algorithm, the actions are discrete values. However, continuous-valued actions can be treated easily by using adaptive PDF [11] or other techniques.…”