Domain Adaptation via Bregman divergence minimization

Authors:

Abstract
In recent years, Fisher linear discriminant analysis (FLDA) based classifi- cation is one of the most successful approaches and has been shown effective performance in different classification tasks. However, when the learning data (source domain) have a different distribution against the testing data (tar- get domain), the FLDA-based models may not be optimal, and the perfor- mance will be degraded, dramatically. To face this problem, in this paper, we propose an optimal domain adaptation via Bregman divergence minimization (DAB) approach, in which the discriminative features of source and target do- mains are simultaneously learned via domain invariant representation. DAB is designed based on the constraints of FLDA, with the objective to adapt the coupled marginal and conditional distribution mismatches with Breg- man divergence minimization. Thus, the resulting representation can have well functionality like FLDA and simultaneously have better discrimination ability. Moreover, our proposed approach can be easily kernelized to deal with nonlinear tasks. Extensive experiments on various benchmark datasets show that our DAB can effectively deal with the cross domain divergence and outperforms several state-of-the-art domain adaptation approaches on cross- distribution domains.