Viet-Anh on Software Logo

What is: GAN Hinge Loss?

SourceGeometric GAN
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

The GAN Hinge Loss is a hinge loss based loss function for generative adversarial networks:

L_D=E_(x,y)p_data[min(0,1+D(x,y))]E_zp_z,yp_data[min(0,1D(G(z),y))]L\_{D} = -\mathbb{E}\_{\left(x, y\right)\sim{p}\_{data}}\left[\min\left(0, -1 + D\left(x, y\right)\right)\right] -\mathbb{E}\_{z\sim{p\_{z}}, y\sim{p\_{data}}}\left[\min\left(0, -1 - D\left(G\left(z\right), y\right)\right)\right]

L_G=E_zp_z,yp_dataD(G(z),y)L\_{G} = -\mathbb{E}\_{z\sim{p\_{z}}, y\sim{p\_{data}}}D\left(G\left(z\right), y\right)