Physical Sciences and Mathematics Commons^™

Adaptive Step-Sizes For Reinforcement Learning, William C. Dabney

Doctoral Dissertations

The central theme motivating this dissertation is the desire to develop reinforcement learning algorithms that “just work” regardless of the domain in which they are applied. The largest impediment to this goal is the sensitivity of reinforcement learning algorithms to the step-size parameter used to rescale incremental updates. Adaptive step-size algorithms attempt to reduce this sensitivity or eliminate the step-size parameter entirely by automatically adjusting the step size throughout the learning process. Such algorithms provide an alternative to the standard “guess-and-check” methods used to find parameters known as parameter tuning. However, the problems with parameter tuning are currently masked by …