Recently, learning discriminative features to improve the recognition performances gradually becomes the primary goal of deep learning, and numerous remarkable works have emerged. In this paper, we propose a novel yet extremely simple method Virtual Softmax to enhance the discriminative property of learned features by injecting a dynamic virtual negative class into the original softmax. Injecting virtual class aims to enlarge inter-class margin and compress intra-class distribution by strengthening the decision boundary constraint. Although it seems weird to optimize with this additional virtual class, we show that our method derives from an intuitive and clear motivation, and it indeed encourages the features to be more compact and separable. This paper empirically and experimentally demonstrates the superiority of Virtual Softmax, improving the performances on a variety of object classification and face verification tasks.
Binghui Chen (Beijing University of Posts and Telecommunications)
Weihong Deng (Beijing University of Posts and Telecommunications)
Haifeng Shen (AI Labs, Didi Chuxing)
Related Events (a corresponding poster, oral, or spotlight)
2018 Poster: Virtual Class Enhanced Discriminative Embedding Learning »
Wed Dec 5th 03:45 -- 05:45 PM Room Room 517 AB