Re:DeepMath Conference 2020 – Conference on the Mathematical...
by aborsy on Saturday November 07, 2020 @01:11PM
(#25017013)
It depends on the function that you are trying to approximate.
I can give you a function that a shallow nnet would approximate better and functions that deep nets approximate exponentially better even with one more layer (in terms of number of neurons n). In the limit n->\infty, both reach arbitrary small errors (obviously often with different number of parameters).
I can give you a function that a shallow nnet would approximate better and functions that deep nets approximate exponentially better even with one more layer (in terms of number of neurons n). In the limit n->\infty, both reach arbitrary small errors (obviously often with different number of parameters).