Deep Representation Learning with Target Coding
AAAI Conference on Artificial Intelligence (AAAI) 2015, Austin, U.S.
Abstract
We consider the problem of learning deep representation when target labels are available. In this paper, we show that there exists intrinsic relationship between target coding and feature representation learning in deep networks. Specifically, we found that distributed binary code with error correcting capability is more capable of encouraging discriminative features, in comparison to the 1-of-K coding that is typically used in supervised deep learning. This new finding reveals additional benefit of using error-correcting code for deep model learning, apart from its well-known error correcting property. Extensive experiments are conducted on popular visual benchmark datasets.
Slides
Citation
@inproceedings{yang2015targetcode,
author = {Shuo Yang, Ping Luo, Chen Change Loy, Kenneth W. Shum and Xiaoou Tang},
title = {Deep Representation Learning with Target Coding},
booktitle = {Proceedings of AAAI Conference on Artificial Intelligence (AAAI)},
month = JAN,
year = {2015}
}