Learning with non-metric proximity matrices
Lists
Only you see this
Abstract
Many emerging applications formulate non-metric proximity matrices (non-positive semidefinite), and hence cannot fit into the framework of kernel machines. A popular approach to this problem is to transform the spectrum of the similarity matrix so as to generate a positive semidefinite kernel matrix. In this paper, we explore four representative transformation methods: denoise, flip, diffusion, and shift. Theoretically, we discuss a generalization problem where the test data are not available during transformation, and thus propose an efficient algorithm to address the problem of updating the cross-similarity matrix between test and training data. Extensive experiments have been conducted to evaluate the performance of these methods on several real-world (dis)similarity matrices with semantic meanings.
BibTeX
@inproceedings{wu2005learning,
author = {Gang Wu and Edward Y. Chang and Zhihua Zhang},
title = {{Learning with non-metric proximity matrices}},
booktitle = {ACM Multimedia},
year = {2005},
url = {https://openreview.net/forum?id=tt8u-D7i4si},
}Loading related content...