Generative-Discriminative Variational Model for Visual Recognition
This work addresses overfitting in visual recognition for researchers and practitioners, but it appears incremental as it builds on existing generative and discriminative approaches without a major breakthrough.
The paper tackles overfitting in deep neural networks for visual recognition by proposing a Generative-Discriminative Variational Model (GDVM) that integrates generative learning with discrimination, showing favorable performance in multi-class, multi-label, and zero-shot learning tasks compared to baselines and recent generative models.
The paradigm shift from shallow classifiers with hand-crafted features to end-to-end trainable deep learning models has shown significant improvements on supervised learning tasks. Despite the promising power of deep neural networks (DNN), how to alleviate overfitting during training has been a research topic of interest. In this paper, we present a Generative-Discriminative Variational Model (GDVM) for visual classification, in which we introduce a latent variable inferred from inputs for exhibiting generative abilities towards prediction. In other words, our GDVM casts the supervised learning task as a generative learning process, with data discrimination to be jointly exploited for improved classification. In our experiments, we consider the tasks of multi-class classification, multi-label classification, and zero-shot learning. We show that our GDVM performs favorably against the baselines or recent generative DNN models.