May 2018
Beginner to intermediate
384 pages
10h 19m
English
Within the context of reinforcement learning, the dynamic programming approach deals with the interactions between a controller or the agent that needs to take actions and the process within the environment. This interaction takes place with three types of distinct signals:
The agent moves through the solution space with repetitive iterations of a state-action-reward-state cycle. A policy defines the overall behavior of the agent. The policy can be dynamically aligned based on the nature of the ...
Read now
Unlock full access