As we have seen in the previous sections, reinforcement learning is a programming philosophy that aims to develop algorithms that can learn and adapt to changes in the environment. This programming technique is based on the assumption of being able to receive stimuli from the outside according to the choices of the algorithm. So, a correct choice will result in a prize while an incorrect choice will lead to a penalization of the system. The goal of the system is to achieve the highest possible prize and consequently the best possible result. The techniques related to learning by reinforcement are divided into two categories:
- Continuous learning algorithms: These techniques start from the assumption of having a simple mechanism able to evaluate the choices of the algorithm and then reward or punish the algorithm depending on the result. These...