Journal
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
Volume 40, Issue 1, Pages 106-118Publisher
IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2017.2666151
Keywords
Metric learning; nearest neighbor classification; f-divergence; generative-discriminative hybridization
Funding
- U.S. Air Force Office of Scientific Research
- National Security Research Institute in Korea
- SNU-MAE BK21+program
- Korea government [IITP-R0126-16-1072, KEIT-10044009, KEIT-10060086]
- U.S. National Science Foundation
- U.S. Department of Transportation
- U.S. Army Research Laboratory
- U.S. Office of Naval Research
Ask authors/readers for more resources
We consider the problem of learning a local metric in order to enhance the performance of nearest neighbor classification. Conventional metric learning methods attempt to separate data distributions in a purely discriminative manner; here we show how to take advantage of information from parametric generative models. We focus on the bias in the information-theoretic error arising from finite sampling effects, and find an appropriate local metric that maximally reduces the bias based upon knowledge from generative models. As a byproduct, the asymptotic theoretical analysis in this work relates metric learning to dimensionality reduction from a novel perspective, which was not understood from previous discriminative approaches. Empirical experiments show that this learned local metric enhances the discriminative nearest neighbor performance on various datasets using simple class conditional generative models such as a Gaussian.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available