I am training a generative adversarial network to perform style transfer from two different image domains (source S and target T). Since I have available class information I have an extra Q network (except G and D) that measures the classification results of the generated images for the target domain and their labels (a LeNet network). From the convergence of the system I have noticed that D is starting always from 8 (the loss function error of the D network) and slightly drops until 4.5 and the G loss function error is starting from 1 and quickly drops to 0.2. Is that behavior an example of mode-collapse? What is exactly the relationship between the errors of D and G? The loss function of D and G I am using can be found here https://github.com/r0nn13/conditional-dcgan-keras while the loss function of Q network is categorical cross-entropy. Is the behavior of D and G normal? Why D loss function is always high?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.