Dear all, I want to put the gradient of neural network (i.e., the gradient of NN output w.r.t input) in the loss function. However, it seems that Zygote cannot differentiate that. Here is my code: using Flux using Zygote using Random Xtrain = rand(Float32,(4,100)); Ytrain = rand(Float32,(1,100));…

How to use gradient of neural network as the loss function?

mkalia94 March 23, 2021, 12:38pm 11

Perhaps @ChrisRackauckas could help out here?

Is it possible to do Nested AD ~elegantly~ in Julia? (PINNs)

Topic		Replies	Views
Flux loss: Gradient wrt input leads to empty gradient wrt parameters or to "can't differentiate foreigncall" Machine Learning flux , forwarddiff , diffeqflux	3	556	April 8, 2022
Gradient error in Flux model inputs Machine Learning question , flux , zygote	5	1324	January 13, 2021
DiffEqFlux Autodifferentiating inside loss function Modelling & Simulations question , diffeq , sciml	6	603	September 29, 2020
Autograd through loss function with derivatives in Machine Learning	1	881	March 25, 2021
Flux differentiation error Machine Learning zygote	19	1684	November 19, 2020