The assumption that training and testing samples are generated from the same distribution does not always hold for real-world machine-learning applications. The procedure of tackling this discrepancy between the training (source) and testing (target) domains is known as domain adaptation. We propose an unsupervised version of domain adaptation that considers the presence of only unlabelled data in the target domain. Our approach centers on finding correspondences between samples of each domain. The correspondences are obtained by treating the source and target samples as graphs and using a convex criterion to match them. The criteria used are first-order and second-order similarities between the graphs as well as a class-based regularization. We have also developed a computationally efficient routine for the convex optimization, thus allowing the proposed method to be used widely. To verify the effectiveness of the proposed method, computer simulations were conducted on synthetic, image classification and sentiment classification datasets. Results validated that the proposed local sample-to-sample matching method outperforms traditional moment-matching methods and is competitive with respect to current local domain-adaptation methods.

No abstract

In this paper, we address the Online Unsupervised Domain Adaptation (OUDA) problem, where the target data are unlabelled and arriving sequentially. The traditional methods on the OUDA problem mainly focus on transforming each arriving target data to the source domain, and they do not sufficiently consider the temporal coherency and accumulative statistics among the arriving target data. We propose a multi-step framework for the OUDA problem, which institutes a novel method to compute the mean-target subspace inspired by the geometrical interpretation on the Euclidean space. This meantarget subspace contains accumulative temporal information among the arrived target data. Moreover, the transformation matrix computed from the mean-target subspace is applied to the next target data as a preprocessing step, aligning the target data closer to the source domain. Experiments on four datasets demonstrated the contribution of each step in our proposed multi-step OUDA framework and its performance over previous approaches.Index Terms-Unsupervised domain adaptation, online domain adaptation, mean subspace, Grassmann manifold.

No abstract

This paper studies mobility strategies to control positions of mobile robots in a mobile sensor network in order to maximize lifetime of the network. With communication and mobility energy costs modeled, the problem is formulated as a nonlinear non-convex optimization problem, then reformulated into a convex optimization problem. The separable property of the system is then exploited by Lagrangian duality, and the solution is obtained by distributed saddle-point computations. Computer simulations showed that the proposed distributed algorithm can quickly converge to the optimal solution, and it also justifies the use of mobility for energy efficiency by showing its significant improvement to the network lifetime and relatively low cost in mobility. Furthermore, the proposed energy optimization framework can accommodate different mobile sensor network models.