Domain Adaptation on Graphs by Learning Aligned Graph Bases

Pilancı, Mehmet
Vural, Elif
A common assumption in semi-supervised learning is that the class label function has a slow variation on the data graph, while in many problems, the label function may vary abruptly in certain graph regions, resulting in high-frequency components. Although semi-supervised learning is an ill-posed problem, it is often possible to find a source graph on which the label function has similar frequency content to the target graph where the actual classification problem is defined. In this paper, we propose a method for domain adaptation on graphs based on learning the spectrum of the label function in a source graph with many labels, and transferring the spectrum information to the target graph. When transferring the frequency content, it is not easy to share graph Fourier coefficients directly between the two independently constructed graphs, since no match exists between their Fourier bases. We solve this by learning a transformation between the Fourier bases of the two graphs that flexibly “aligns” them. The unknown class label function on the target graph is then reconstructed from the learnt spectrum while retaining consistency with the available labels. Comparative experiments suggest that the proposed algorithm often outperforms recent domain adaptation methods in various data classification applications.
IEEE Transactions On Knowledge And Data Engineering


Differential quadrature solution of nonlinear reaction-diffusion equation with relaxation-type time integration
Meral, G.; Tezer, Münevver (Informa UK Limited, 2009-01-01)
This paper presents the combined application of differential quadrature method (DQM) and finite-difference method (FDM) with a relaxation parameter to nonlinear reaction-diffusion equation in one and two dimensions. The polynomial-based DQM is employed to discretize the spatial partial derivatives by using Gauss-Chebyshev-Lobatto points. The resulting system of ordinary differential equations is solved, discretizating the time derivative by an explicit FDM. A relaxation parameter is used to position the sol...
Estimation in bivariate nonnormal distributions with stochastic variance functions
Tiku, Moti L.; İslam, Muhammed Qamarul; SAZAK, HAKAN SAVAŞ (Elsevier BV, 2008-01-01)
Data sets in numerous areas of application can be modelled by symmetric bivariate nonnormal distributions. Estimation of parameters in such situations is considered when the mean and variance of one variable is a linear and a positive function of the other variable. This is typically true of bivariate t distribution. The resulting estimators are found to be remarkably efficient. Hypothesis testing procedures are developed and shown to be robust and powerful. Real life examples are given.
Estimation and hypothesis testing in BIB design and robustness
Tiku, Moti L.; ŞENOĞLU, BİRDAL (Elsevier BV, 2009-07-01)
Modified maximum likelihood estimators of the unknown parameters in a BIB design under non-normality of error distributions are obtained. They are shown to be more efficient and robust than the traditional least squares estimators. A test statistic for testing a linear contrast among treatment effects is developed. A real life example is given.
Sampling of the Wiener Process for Remote Estimation Over a Channel With Random Delay
Sun, Yin; Polyanskiy, Yury; Uysal, Elif (Institute of Electrical and Electronics Engineers (IEEE), 2020-02-01)
In this paper, we consider a problem of sampling a Wiener process, with samples forwarded to a remote estimator over a channel that is modeled as a queue. The estimator reconstructs an estimate of the real-time signal value from causally received samples. We study the optimal online sampling strategy that minimizes the mean square estimation error subject to a sampling rate constraint. We prove that the optimal sampling strategy is a threshold policy, and find the optimal threshold. This threshold is determ...
Random Set Methods Estimation of Multiple Extended Objects
Granstrom, Karl; Lundquist, Christian; Gustafsson, Fredrik; Orguner, Umut (Institute of Electrical and Electronics Engineers (IEEE), 2014-06-01)
Random set-based methods have provided a rigorous Bayesian framework and have been used extensively in the last decade for point object estimation. In this article, we emphasize that the same methodology offers an equally powerful approach to estimation of so-called extended objects, i.e., objects that result in multiple detections on the sensor side. Building upon the analogy between Bayesian state estimation of a single object and random finite set (RFS) estimation for multiple objects, we give a tutorial...
Citation Formats
M. Pilancı and E. Vural, “Domain Adaptation on Graphs by Learning Aligned Graph Bases,” IEEE Transactions On Knowledge And Data Engineering, pp. 1–14, 2020, Accessed: 00, 2020. [Online]. Available: