The final, trained weights of neural network solution are highly reproducible across multiple instances of neural network optimized for the same problem while starting from different random sets of initial values, and the weights are directly interpretable in terms of relative importance of the input variables

Group of answer choices

True

False