| 摘要: |
| 针对现有的基于图的半监督学习（graph-based semi-supervised learning，简称GSSL）方法存在模型参数敏感和数据空间判别信息不充分等问题，受最近特征空间嵌入和数据稀疏表示思想的启发，提出一种稀疏近似最近特征空间嵌入标签传播算法SANFSP（sparse approximated nearest feature space embedding label propagation）.SANFSP首先利用特征空间嵌入投影点来稀疏表示原始数据；然后，度量原始数据和稀疏近似最近特征空间嵌入投影间的相似性；进而提出稀疏近似最近特征空间嵌入正则化项；最后，基于传统GSSL 方法的标签传播算法，实现数据标签的平滑传播.同时，还将SANFSP 算法简单拓展到out-of-sample 学习.SANFSP 算法在人造和实际数据集（如人脸识别、可视物件识别以及手写数字分类等）上取得了有效的实验结果. |
| 关键词: 半监督学习 稀疏表示 标签传播 最近特征空间嵌入 |
| DOI：10.13328/j.cnki.jos.004556 |
| 分类号: |
| 基金项目:教育部人文社会科学研究规划基金（13YJAZH084）；浙江省自然科学基金（LY14F020009，LY13F020011）；宁波市自然科学基金（2013A610065，2013A610072）；香港理工大学基金（G-UA68） |
|
| Label Propagation Using Sparse Approximated Nearest Feature Space Embedding |
|
TAO Jian-Wen1, Fu-Lai CHUNG2, WANG Shi-Tong2,3, YAO Qi-Fu4
|
|
1.School of Information Science and Engineering, Ningbo Institute of Technology, Zhejiang University, Ningbo 315100, China;2.Department of Computing, Hong Kong Polytechnic University, Hong Kong, China;3.School of Digital Media, Jiangnan University, Wuxi 214122, China;4.School of Information Engineering, Zhejiang Business Technology Institute, Ningbo 315012, China
|
| Abstract: |
| There exist several problems in existing graph-based semi-supervised learning (GSSL) methods such as model parameters sensitiveness and insufficient discriminative information in data space, etc. To address those issues, this paper proposes a sparse approximated nearest feature space embedding label propagation (SANFSP) algorithm, which is inspired by both ideas of nearest feature space embedding and that of sparse representation. SANFSP first sparsely reconstructs data from original space using its feature space embedding projection images, and then measures the similarity between original data and its sparse approximated nearest feature space embedding projection points, thus proposing a sparse approximated nearest feature space embedding regularizer. At last, SANFSP complets label propagation procedure by using classical label propagation algorithm. The study also derives an easy way to extend SANFSP to out-of-sample data. Promising experimental results are obtained on several toy and real-world classification tasks such as face recognition, visual object recognition and digit classification. |
| Key words: semi-supervised learning sparse representation label propagation nearest feature space |
