Yes and what’s more is that natural selection is a loss minimization function, similar to gradient descent