April 2018
Intermediate to advanced
334 pages
10h 18m
English
In this book, we have covered most of the algorithms in the area of reinforcement learning from basic to advanced. Therefore, those chapters are prerequisites to understand applications and challenges faced by different algorithms in the domain of robotics. Early reinforcement learning algorithms dealt in obtaining optimal policies by first obtaining state action values and then deriving the policy from them. Then, policy iteration methods came into the picture, which are directly used to output the optimized policy. The exploration-exploitation techniques helped in refining existing policies, exploring new actions, and updating the existing policies. Reinforcement learning approaches, such as MDP (in ...