Cross-Domain Sparse Coding
This addresses the challenge of learning across domains with different distributions, which is important for applications like image classification and spam detection, though it appears to be an incremental extension of sparse coding.
The paper tackles the problem of extending sparse coding to cross-domain learning where source and target domains have significantly different distributions, achieving improved performance on image classification and spam detection tasks compared to other cross-domain representation methods.
Sparse coding has shown its power as an effective data representation method. However, up to now, all the sparse coding approaches are limited within the single domain learning problem. In this paper, we extend the sparse coding to cross domain learning problem, which tries to learn from a source domain to a target domain with significant different distribution. We impose the Maximum Mean Discrepancy (MMD) criterion to reduce the cross-domain distribution difference of sparse codes, and also regularize the sparse codes by the class labels of the samples from both domains to increase the discriminative ability. The encouraging experiment results of the proposed cross-domain sparse coding algorithm on two challenging tasks --- image classification of photograph and oil painting domains, and multiple user spam detection --- show the advantage of the proposed method over other cross-domain data representation methods.