Knowledge Commons of Institute of Automation,CAS
Integrating supervised subspace criteria with restricted Boltzmann Machine for feature extraction | |
Xie, Guo-Sen; Zhang, Xu-Yao; Zhang, Yan-Ming; Liu, Cheng-Lin | |
2014 | |
会议名称 | International Joint Conference on Neural Networks (IJCNN) |
会议录名称 | Procceding of IJCNN 2014 |
会议日期 | 2014 |
会议地点 | 北京 |
摘要 | Restricted Boltzmann Machine (RBM) is a widely used building-block in deep neural networks. However, RBM is an unsupervised model which can not exploit the rich supervised information of data. Therefore, we consider combining the descriptive (generative) ability of RBM with the discriminative ability of supervised subspace models, i.e., Fisher linear discriminant analysis (FDA), marginal Fisher analysis (MFA), and heat kernel MFA (hkMFA). Specifically, the hidden layer of RBM is regularized by the supervised subspace criteria, and the joint learning model can then be efficiently optimized by gradient descent and graph construction (used to define the scatter matrix in the subspace models) on mini-batch data. Compared with the traditional subspace models (FDA, MFA, hkMFA), the proposed hybrid models are essentially nonlinear and can be optimized by gradient descent instead of eigenvalue decomposition. More importantly, traditional subspace models can only reduce the dimensionality (because of linear transformation), while the proposed models can also increase the dimensionality for better class discrimination. Experiments on three databases demonstrate that the proposed hybrid models outperform both RBM and their counterpart subspace models (FDA, MFA, hkMFA) consistently. |
关键词 | Rbm |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/11958 |
专题 | 多模态人工智能系统全国重点实验室_模式分析与学习 |
通讯作者 | Xie, Guo-Sen |
推荐引用方式 GB/T 7714 | Xie, Guo-Sen,Zhang, Xu-Yao,Zhang, Yan-Ming,et al. Integrating supervised subspace criteria with restricted Boltzmann Machine for feature extraction[C],2014. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
IJCNN14_paper.pdf(494KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论