I am trying to add a gradient norm penalty to a loss function similar to WGAN-GP ( [1704.00028] Improved Training of Wasserstein GANs , Alg 1, Lines 7-9). However, I am running into problems with mutation. Is there another way to do this, or does something have to change in Zygote to accomodate this u…

How to add norm of gradient to a loss function?

ToucheSir October 27, 2021, 12:07am 13

As always, please include a full stacktrace along with your MWE.

Topic		Replies	Views
How to do L2 regularization with new Flux and Zygote Machine Learning flux	2	2116	December 29, 2019
Clipping gradients with Zygote/Flux Machine Learning	2	1183	June 10, 2019
M. learning with regularization using Flux is too slow? Performance question , flux	5	398	February 9, 2024
Gradient error in Flux model inputs Machine Learning question , flux , zygote	5	1300	January 13, 2021
How to use gradient of neural network as the loss function? Machine Learning question	13	2703	March 23, 2021