Amir-Arsalan/Synthesize3DviaDepthOrSil

Question about the loss

DavisonHu opened this issue · 2 comments

Hello,I read the paper "Synthesizing 3D Shapes via Modeling Multi-view Depth Maps and Silhouettes with Deep Generative Networks",the loss function which mention inside is to minimize : loss = -DKL(q(Z|X)||p(Z)) + E[logp(x|Z)]
but the loss function in VAE is :
loss = DKL(q(Z|X)||p(Z)) - E[logp(x|Z)]
I am confused about it, I hope you can give me some advice,thank you

They are essentially the same. You always optimize the parameters by going in the negative direction of the gradients, hence we put the negative sign there.

@DaviosnHu This will clarify things for you.