Generative Ratio Matching Networks
This work addresses the training instability problem for researchers and practitioners in generative modeling, offering a stable alternative to adversarial methods.
The paper tackles the instability of adversarial training in deep generative models by introducing Generative Ratio Matching (GRAM), a method that uses a fixed kernel as an adversary to avoid saddlepoint optimization. GRAM achieves stable training and matches or exceeds the generative quality of adversarial networks.
Deep generative models can learn to generate realistic-looking images, but many of the most effective methods are adversarial and involve a saddlepoint optimization, which requires a careful balancing of training between a generator network and a critic network. Maximum mean discrepancy networks (MMD-nets) avoid this issue by using kernel as a fixed adversary, but unfortunately, they have not on their own been able to match the generative quality of adversarial training. In this work, we take their insight of using kernels as fixed adversaries further and present a novel method for training deep generative models that does not involve saddlepoint optimization. We call our method generative ratio matching or GRAM for short. In GRAM, the generator and the critic networks do not play a zero-sum game against each other, instead, they do so against a fixed kernel. Thus GRAM networks are not only stable to train like MMD-nets but they also match and beat the generative quality of adversarially trained generative networks.