Sharif Digital Repository

Learning an appropriate distance measure under supervision of side information has become a topic of significant interest within machine learning community. In this paper, we address the problem of metric learning for constrained clustering by considering three important issues: (1) considering importance degree for constraints, (2) preserving the topological structure of data, and (3) preserving some natural distribution properties in the data. This work provides a unified way to handle different issues in constrained clustering by learning an appropriate distance measure. It has modeled the first issue by injecting the importance degree of constraints directly into an objective function.... 

A new, robust homogenization scheme for determination of the effective properties of a periodic piezoelectric composite with general multi-coated inhomogeneities is developed. In this scheme the coating does not have to be thin, the shape and orientation of the inclusion and coatings do not have to be identical, their centers do not have to coincide, their properties do not have to remain uniform, and the microstructure can be with the 2D elliptic or the 3D ellipsoidal inclusions. The development starts from the local electromechanical equivalent inclusion principle through the introduction of the position-dependent equivalent eigenstrain and electric field. Then with a Fourier series...