Summary
In this chapter, we learned how to develop models based on ANNs to simulate physical phenomena. We started by analyzing the basic concepts of neural networks and the principles they are based on that are derived from biological neurons. We examined, in detail, the architecture of an ANN, understanding the concepts of weights, bias, layers, and the activation function.
Subsequently, we analyzed the architecture of a feedforward neural network. We saw how the training of the network with data takes place, and we understood the weight adjustment procedure that leads the network to correctly recognize new observations.
Next, we applied the concepts learned by tackling a practical case. We developed a model based on neural networks to solve a regression problem. We learned how to scale data and then how to subset the data for training and testing. We learned how to develop a model based on linear and MLP regression and how to evaluate the performance of these models to make...
