Max-Margin Deep Generative Models for (Semi-)Supervised Learning
This work addresses the challenge of enhancing discriminative ability in DGMs for researchers and practitioners in machine learning, representing an incremental improvement by integrating max-margin learning into existing frameworks.
The paper tackles the problem of improving the predictive performance of deep generative models (DGMs) while retaining their generative ability, by introducing max-margin deep generative models (mmDGMs) and a class-conditional variant (mmDCGMs). The results show that mmDGMs are competitive with fully discriminative networks in supervised learning and achieve state-of-the-art classification results in semi-supervised learning on several benchmarks.
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, it is relatively insufficient to empower the discriminative ability of DGMs on making accurate predictions. This paper presents max-margin deep generative models (mmDGMs) and a class-conditional variant (mmDCGMs), which explore the strongly discriminative principle of max-margin learning to improve the predictive performance of DGMs in both supervised and semi-supervised learning, while retaining the generative capability. In semi-supervised learning, we use the predictions of a max-margin classifier as the missing labels instead of performing full posterior inference for efficiency; we also introduce additional max-margin and label-balance regularization terms of unlabeled data for effectiveness. We develop an efficient doubly stochastic subgradient algorithm for the piecewise linear objectives in different settings. Empirical results on various datasets demonstrate that: (1) max-margin learning can significantly improve the prediction performance of DGMs and meanwhile retain the generative ability; (2) in supervised learning, mmDGMs are competitive to the best fully discriminative networks when employing convolutional neural networks as the generative and recognition models; and (3) in semi-supervised learning, mmDCGMs can perform efficient inference and achieve state-of-the-art classification results on several benchmarks.