Reinforcement Learning is one of the most interesting fields of Machine Learning. The software acts according to observations made in an environment, as a result, it gets rewards. Its target is to learn actions leading to maximizing expected rewards in the long term. In short, the agent acts in the environment and learns by trial and error to maximize rewards and minimize sanctions.
The main characteristic is that, due to the reward-sanction based system, the software learns from the mistakes made in every action and improves the method´s quality. Similarly, a machine can make decisions even if it does not store any a priori knowledge of the environment or variables and make more advanced abstract matters satisfactorily.
Grupo AIA is leader in the use of the most recent algorithms and models such as Reinforcement Learning in areas such as:
DNA sequence classification
Gas Distribution Optimization