Timezone: »

 
Realizing GANs via a Tunable Loss Function
Gowtham Raghunath Kurri · Tyler Sypherd · Lalitha Sankar

@
in Information-Theoretic Methods for Rigorous, Responsible, and Reliable Machine Learning (ITR3) »
We introduce a tunable GAN, called $\alpha$-GAN, parameterized by $\alpha \in (0,\infty]$, which interpolates between various $f$-GANs and Integral Probability Metric based GANs (under constrained discriminator set). We construct $\alpha$-GAN using a supervised loss function, namely, $\alpha$-loss, which is a tunable loss function capturing several canonical losses. We show that $\alpha$-GAN is intimately related to the Arimoto divergence, which was first proposed by \"{O}sterriecher (1996), and later studied by Liese and Vajda (2006). We posit that the holistic understanding that $\alpha$-GAN introduces will have practical benefits of addressing both the issues of vanishing gradients and mode collapse.

Author Information

Gowtham Raghunath Kurri (Arizona State University)
Tyler Sypherd (Arizona State University)
Lalitha Sankar (Arizona State University)

More from the Same Authors