What kind of a neural network will you use in deep learning regression via Keras-TensorFlow? Or How will you decide the best neural network model for a given problem?

The most important step in deciding on a neural network model is knowing your data well and choosing the best model. Also, it is important to consider whether the problem is linearly separable when deciding on a neural network model. As such, the task and data at hand play an important role in choosing the best neural network model for a given problem. However, unlike CNNs, LSTMs, or RNNs, which require a composition of nodes and layers, it is always recommended to start with simple models like Multilayer Perceptrons (MLP), which have only one hidden layer. Therefore, MLP is considered the simplest neural network. This is because weight initialization is unimportant, and we do not need to define the network structure in advance.