Max-margin Deep Generative Models

Deep generative models (DGMs) are effective on learning multilayered representations of complex data and performing inference of input data by exploring the generative ability. However, little work has been done on examining whether the representations are discriminative enough to get good prediction performance. In this paper, we present max-margin deep generative models (mmDGMs), which explore the strongly discriminative principle of max-margin learning to improve the discriminative power of DGMs, while retaining the generative capability. We develop an efficient doubly stochastic subgradient algorithm. Empirical results on MNIST and its variant datasets demonstrate that (1) max-margin learning can significantly improve the classification performance of DGMs and meanwhile retain the ability of inferring input data; and (2) mmDGMs are competitive to the state-of-the-art networks that have a similar structure.