Timezone: »

 
Poster
Learning Kernels with Radiuses of Minimum Enclosing Balls
Kun Gai · Guangyun Chen · Changshui Zhang

Mon Dec 06 12:00 AM -- 12:00 AM (PST) @
in Poster Session and Reception »

In this paper, we point out that there exist scaling and initialization problems in most existing multiple kernel learning (MKL) approaches, which employ the large margin principle to jointly learn both a kernel and an SVM classifier. The reason is that the margin itself can not well describe how good a kernel is due to the negligence of the scaling. We use the ratio between the margin and the radius of the minimum enclosing ball to measure the goodness of a kernel, and present a new minimization formulation for kernel learning. This formulation is invariant to scalings of learned kernels, and when learning linear combination of basis kernels it is also invariant to scalings of basis kernels and to the types (e.g., L1 or L2) of norm constraints on combination coefficients. We establish the differentiability of our formulation, and propose a gradient projection algorithm for kernel learning. Experiments show that our method significantly outperforms both SVM with the uniform combination of basis kernels and other state-of-art MKL approaches.

[ PDF] 

Author Information

Kun Gai (Alibaba group)
Guangyun Chen (Tsinghua University)
Changshui Zhang (Tsinghua University)

More from the Same Authors