Key questions for the artificial neural network learner to ask themselves

Over the recent years, Artificial Neural Networks (ANNs) have significantly matured, yet numerous unresolved issues persist. Instead of reviewing these open questions, this perspective piece will present my viewpoint on how to tackle problems within the field of ANNs. Specifically, I will outline a set of critical questions we should consider when developing ANN algorithms for machine learning. These questions will center on the fundamental definition of ANNs and their connection to the human brain, the processes of forward and backward propagation, the concepts of training, optimization, and iteration, the methods for minimizing the discrepancy between predicted and actual values, and the recent advancements in model architecture and design.