Python Natural Language Processing

Reinforcement learning

The third type of ML is reinforcement learning. Here, the ML algorithm doesn't give you the feedback right after every prediction, but it generates feedback if the ML model achieves its goal. This type of learning is mostly used in the area of robotics and to develop intelligent bots to play games. Reinforcement learning is linked to the idea of interacting with an environment using the trial and error method. Refer to Figure 8.3:

Figure 8.3: Reinforcement learning interacting with environment (Image credit: https://devblogs.nvidia.com/parallelforall/wp-content/uploads/2016/04/aeloop-300x183.png)

To learn the basics, ...