Reinforcement Learning
Reinforcement LearningIt is an important branch of machine learning and an interdisciplinary product. Its essence is to solve the decision making problem, that is, to achieve automatic decision-making and continuous decision-making.
Reinforcement learning mainly includes four elements: Agent, environment state, behavior, and reward. Its goal is to obtain the most cumulative rewards.
Reinforcement Learning Classification
From the perspective of elements, there are mainly the following methods:
- Policy based: The focus is on finding the optimal policy;
- Value based: The focus is on finding the optimal sum of rewards;
- Action based: The focus is on the optimal action at each step.