A model is only useful based on its predictive power versus its complexity. The math of physics is about providing the simplest model possible to explain what we observe in reality. A neural network or any universal approximating function should be able to describe the entirety of reality with enough parameters. Lots and lots of different models can describe the same thing but no model is right. Every model is wrong. Models are really only useful or not.
First a watch, now a neural network. Nothing new under the sun.<p>> All that is needed is to find a physical phenomenon which cannot be described by neural networks.<p>But neural networks are universal function approximations...<p>Does this offer any falsifiable predictions? I freely admit that my understanding of physics at this level is superficial - can anyone offer insight on the value of this paper?