Table 3 Comparison of different reinforcement machine learning methods
From: Healthcare predictive analytics using machine learning and deep learning techniques: a survey
Method | Advantages | Disadvantages |
|---|---|---|
Q-learning | • It can be applied to a wide range of problems, including those that are difficult to solve using other methods • It can learn from experience, which means it can improve its performance over time • It is relatively simple to implement and can be used in a variety of settings | • It can be difficult to learn, particularly for problems with large state spaces • It can be sensitive to initial conditions, which means that if it is not properly initialized, it may learn a suboptimal policy |
Monte Carlo tree search | • It is useful for solving problems with a large search space • It can learn from experience, which means it can improve its performance over time • It is relatively simple to implement and can be applied to a wide range of problems | • It can be difficult to learn, particularly for problems with a large search space • It can be sensitive to initial conditions, which means that if it is not properly initialized, it may learn a suboptimal policy • Due to the requirement to store a tree of possible states, it can be difficult to scale to large problems |