Understanding RL
Compared to the various forms of AI, RL is characterized by implementing a trial and error fashion of automated learning. In fact, the RL algorithms adapt their learning processes based on the feedback obtained from the environment. This feedback can be positive, that is, rewards; or negative, that is, punishments. In addition, feedback differs according to the successes and errors of the predictions.
Therefore, we can say that learning takes place on the basis of rewards and punishments obtained by an intelligent software: as such, the intelligent software (also known as the agent) learns from the feedbacks obtained from a given domain contest (also known as the environment).
Unlike ML, in RL the learning process does not ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access