The final type of ML is reinforced learning. This is a process where the machine learns by itself based on a reward system. In reinforced learning, there is generally an environment and an agent, and the agent is typically the machine. Let's say that the environment is in state A and an agent performs action A on this environment and observes how the state changes. Based on the action, the new state, and the reward returned by the environment, the machine learns.
Think of reinforced learning as a machine learning how to ride a bike. If the machine is able to move forward, it receives a reward, and if the machine falls down because it lost its balance, it is not rewarded.
Reinforced learning forms the basis of a lot of AI systems.
If you wish to learn more about machine learning, please refer to the following resources:
- Python Machine Learning - Second Edition: https://www.packtpub.com/big-data-and-business-intelligence/python-machine-learning-second-edition
- Python Deep Learning: https://www.packtpub.com/big-data-and-business-intelligence/python-deep-learning
- Practical Machine Learning: https://www.packtpub.com/big-data-and-business-intelligence/practical-machine-learning