Multi-Task Learning in Heterogeneous Feature Spaces

By Yu Zhang, Dit-Yan Yeung. 2011

65 citations recorded by the source corpus.

Multi-task learning aims at improving the generalization performance of a learning task with the help of some other related tasks. Although many multi-task learning methods have been proposed, they are all based on the assumption that all tasks share the same data representation. This assumption is too restrictive for general applications. In this paper, we propose a multi-task extension of linear discriminant analysis (LDA), called multi-task discriminant analysis (MTDA), which can deal with learning tasks with different data representations. For each task, MTDA learns a separate transformation which consists of two parts, one specific to the task and one common to all tasks. A by-product of MTDA is that it can alleviate the labeled data deficiency problem of LDA. Moreover, unlike many existing multi-task learning methods, MTDA can handle binary and multi-class problems for each task in a generic way. Experimental results on face recognition show that MTDA consistently outperforms related methods.

Subfields: Artificial Intelligence, Computational Mechanics, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Dacheng Tao
2. J. Lai
3. Weishi Zheng
4. Jianfang Hu
5. Jianguo Zhang
6. Tongliang Liu
7. Yong Luo
8. Yuxin Peng

In Computational Mechanics

1. Tongliang Liu
2. Dacheng Tao
3. Yong Luo
4. Yonggang Wen
5. Xiaohua Zhai
6. Jianguo Zhang
7. Weishi Zheng
8. J. Lai

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Yuxin Peng
3. J. Lai
4. Weishi Zheng
5. Xiaogang Wang
6. G. Wang
7. Luc Van Gool
8. Yi Yang

Citations in the corpus

• General Heterogeneous Transfer Distance Metric Learning via Knowledge Fragments Transfer (2017, methodology): “The hyper-parameter of intermediate dimensionality is set as a fixed value since the model is not very sensitive to it according to [Zhang and Yeung, 2011].”
• Structured Images for RGB-D Action Recognition (2017, unknown): “22% DS+DCP+DDP+MTDA [42] 84.”
• Jointly Learning Heterogeneous Features for RGB-D Activity Recognition (2015, methodology): “In addition, our JOULE works better than MTDA on CAD 60 set with a smaller performance gain than on the MSR Daily set.”
• Heterogeneous Metric Learning with Joint Graph Regularization for Cross-Media Retrieval (2013, background): “Recently, learning from heterogeneous data has shown effectiveness in heterogeneous transfer learning(Zhu et al. 2011), heterogeneous multi task learning(Zhang and Yeung 2011), and transductive classification on heterogeneous information networks(Ji et al. 2010).”