In this chapter, we explored the machine learning field and we saw the learning process in a neural network. We learned to distinguish between supervised learning, unsupervised learning, and reinforcement learning. To understand in detail the necessary procedures, we also learned how to train and test the model.
Afterwards, we discovered the meaning of the data cycle and how the data must be collected, cleaned, converted, and then fed to the model for learning. So we went deeper into the evaluation model to see if the expected value is equal to the actual value during the test phase. We analyzed the different metrics available to control the model that depends on the status of the target variable.
Then we discovered one of the concepts important for understanding the neural networks, the backpropagation algorithm, that is based on computing to update weights and bias ions at each level.
Finally, we covered two practical programs in R for the learning process, by applying the neuralnet...