Geometry-aware Deep Transform
This addresses the challenge of limited labeled data in many applications, though it appears incremental as it builds on existing deep learning and metric learning methods.
The paper tackles the problem of poor generalization in deep learning with small training sets by proposing a novel objective that unifies classification and metric learning, resulting in competitive performance on both synthetic and real-world data.
Many recent efforts have been devoted to designing sophisticated deep learning structures, obtaining revolutionary results on benchmark datasets. The success of these deep learning methods mostly relies on an enormous volume of labeled training samples to learn a huge number of parameters in a network; therefore, understanding the generalization ability of a learned deep network cannot be overlooked, especially when restricted to a small training set, which is the case for many applications. In this paper, we propose a novel deep learning objective formulation that unifies both the classification and metric learning criteria. We then introduce a geometry-aware deep transform to enable a non-linear discriminative and robust feature transform, which shows competitive performance on small training sets for both synthetic and real-world data. We further support the proposed framework with a formal $(K,ε)$-robustness analysis.