Earticle

초록

영어

Multi-task Learning (MTL) algorithms aim to improve the performance of several learning methods through shared information among all tasks. One particularly successful instance of multi-task learning is its adaptation to support vector machine (SVM). Recently advances in large-margin learning have shown that their solutions may be misled by the spread of data and preferentially separate classes along large spread directions. In this paper, we propose a novel formulation for multi-task learning by extending the recently published relative margin machine algorithm to the multi-task learning paradigm. The new method is an extension of support vector machine for single task learning. The objective of our algorithm is to obtain a different predictor for each task while taking into account the fact that the tasks are related as well as the spread of the data. We test the proposed method experimentally using real data. The experiments show that the proposed method performs better than existing multi-task leaning with SVM and single-task leaning with SVM.

목차

Abstract
 1. Introduction
 2. Relative Margin Machine
 3. Relative Margin Multi-Task Learning (RMMTL)
  3.1. Linear Relative Margin Multi-Task Learning
  3.2. Nonlinear Relative Margin Multi-Task Learning
 4. Experiments
  4.1. Dermatology Dataset
  4.2. Isolet Dataset
  4.3. Monk Dataset
  4.4. Radar Landmine Detection Dataset
 5. Conclusions and Discussion
 Acknowledgments
 References

키워드

저자

• Yunyan Song [ College of Science, Tianjin University of Technology, Tianjin, 300384, China ]
• Wenxin Zhu [ College of Basic Science, Tianjin Agricultural University, Tianjin, 300384, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Signal Processing, Image Processing and Pattern Recognition

• 간기

  격월간

• pISSN

  2005-4254

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605