Ángel Rodríguez-Rozas
Ph.D. student at the Center for Mathematics and its Applications (CEMAT), department of Mathematics, Instituto Superior Técnico.

	Personal data: Birth date: 30.11.1983 Nationality: Spain Contact e-mail: Phone: (+351) 218 417 883 FAX: (+351) 218 417 048 Address: Instituto Superior Tecnico Department of Mathematics Av. Rovisco Pais 1049-001. Lisboa. Portugal
UC Berkeley campus, California, US, summer 2011.

Ángel Rodríguez-Rozas is currently a Ph.D. student in Computational Science and Engineering. He pursued his studies in Computer Science at University of Rovira I Virgili of Tarragona, receiving the outstanding graduation award.

His current research interests focus on the development of new efficient numerical algorithms for high performance scientific computing. This research field is usually referred to as Computational Mathematics, and may be considered as a meeting point between Computational Physics, Nonlinear Physics, Applied Mathematics, Numerical Mathematics, Computational Science and Engineering, Scientific Computing, and Parallel Computing. The common denominator is the fact of making realistic predictions on challenging problems arising from Science and Engineering. Such a problems, being nonlinear, usually requires a huge computational effort to be solved. Despite Supercomputers have already entered in the petascale computing era, at the present time almost all algorithms are not capable of efficiently exploit such a computational resources mainly due to the well-know parallel computing related-issues.

In his Ph.D. thesis under the supervision of Dr. Juan A. Acebrón, he is developing new algorithms based on a probablisitc domain decomposition for solving partial differential equations. Such algorithm, called PDD (Probabilistic Domain Decomposition), being based on Monte Carlo techniques, has shown its capabilities to overcome the inherently synchronization and communication limitations that classical algorithms usually have, being the original problems fully decoupled into as many subproblems as the number of available processors, exhibiting a remarkable scalability. Such a method is being extended to solve a wide range of nonlinear partial differential equations, whose solution can be represented probabilistically being amenable solved by the PDD algorithm. Among them it is being considering very important equations, such as the Maxwell-Vlasov, and the Navier-Stokes equations, governing Plasma Physics, and Fluid dynamics problems, respectively.

He has also had general interests and contributions in Artificial Intelligence during the bachelor's degree, specifically in the Multi-Agent System paradigm and their capabilities for general purpose parallel computing, with particular application in complex knowledge acquisition methods.