DAML: Domain Adaptation Metric Learning

Authors:
Bo Geng;Dacheng Tao;Chao Xu
Affiliations:
Key Laboratory of Machine Perception (Ministry of Education), Peking University, Beijing, China;Centre for Quantum Computation and Intelligent Systems, Faculty of Engineering and Information Technology, University of Technology, Sydney, Broadway, Australia;Key Laboratory of Machine Perception (Ministry of Education), Peking University, Beijing, China
Venue:
IEEE Transactions on Image Processing
Year:
2011

Citing 0
Cited 12

Dimensionality reduction by Mixed Kernel Canonical Correlation Analysis

Pattern Recognition
CIPCA: Complete-Information-based Principal Component Analysis for interval-valued data

Neurocomputing
Extracting elite pairwise constraints for clustering

Neurocomputing
LF-EME: Local features with elastic manifold embedding for human action recognition

Neurocomputing
Exploiting visual word co-occurrence for image retrieval

Proceedings of the 20th ACM international conference on Multimedia
Generalization performance of magnitude-preserving semi-supervised ranking with graph-based regularization

Information Sciences: an International Journal
Locally regularized sliced inverse regression based 3D hand gesture recognition on a dance robot

Information Sciences: an International Journal
Cartoon features selection using Diffusion Score

Signal Processing
Learning colours from textures by sparse manifold embedding

Signal Processing
Asymmetric propagation based batch mode active learning for image retrieval

Signal Processing
Similar handwritten Chinese character recognition by kernel discriminative locality alignment

Pattern Recognition Letters
Sparse Discriminative Information Preservation for Chinese character font categorization

Neurocomputing

Quantified Score

Hi-index	0.01

Visualization

Abstract

The state-of-the-art metric-learning algorithms cannot perform well for domain adaptation settings, such as cross-domain face recognition, image annotation, etc., because labeled data in the source domain and unlabeled ones in the target domain are drawn from different, but related distributions. In this paper, we propose the domain adaptation metric learning (DAML), by introducing a data-dependent regularization to the conventional metric learning in the reproducing kernel Hilbert space (RKHS). This data-dependent regularization resolves the distribution difference by minimizing the empirical maximum mean discrepancy between source and target domain data in RKHS. Theoretically, by using the empirical Rademacher complexity, we prove risk bounds for the nearest neighbor classifier that uses the metric learned by DAML. Practically, learning the metric in RKHS does not scale up well. Fortunately, we can prove that learning DAML in RKHS is equivalent to learning DAML in the space spanned by principal components of the kernel principle component analysis (KPCA). Thus, we can apply KPCA to select most important principal components to significantly reduce the time cost of DAML. We perform extensive experiments over four well-known face recognition datasets and a large-scale Web image annotation dataset for the cross-domain face recognition and image annotation tasks under various settings, and the results demonstrate the effectiveness of DAML.