Dos Reis receives NSF CAREER Award

Dr. Gabriel Dos Reis, assistant professor in the Department of Computer Science and Engineering, received the National Science Foundation (NSF) CAREER award for his continuing investigation of formal-methods-based principles and tools to make the practice of programming a more mathematical activity for ordinary programmers.

The NSF awards the prestigious CAREER grants to outstanding junior faculty members to help them advance their research and teaching activities. His project, "Compilers for Dependable Computational Mathematics," will continue through May 2017.

"The overriding aim is to make dependability a common basic property of criticalPhoto of Dr. Gabriel Dos Reissoftware...The results are expected to influence the evolution of existing major mainstream programming languages (such as C++) in their support for structured generic programming and improvement of software reliability," Dos Reis said.

Our modern civilization has become extraordinarily dependent on software artifacts at virtually all levels of its structure. An immediate implication is that failures in software systems can cost human lives and are responsible for billions of dollars lost annually. Furthermore, silent failures of simulation software can seriously undermine the credibility of a scientific model, entailing profound intellectual liabilities. On the other hand, the algorithms we use in computational mathematics are becoming more and more complex. Yet the programming languages we use to express them as computer programs do not seem to progress at a matching pace. As a result, there is an ever growing gap between the level of abstraction of published algorithms, and the software artifacts that realize them. In other words, the mapping between algorithms and code is no longer direct nor obvious, to the point where we have trouble convincing ourselves that our programs really do what they are supposed to do.

Dos Reis's research into axiomatic programming rests upon the philosophy that large scale dependable mathematics software should be developed in high level programming languages that enable direct expression of algorithms that ideally are in obvious correspondence with their mathematical formulations. He will investigate and develop tools to support research in axiomatic programming for computational mathematics. He will design new programming models rooted in a structured approach to generic programming. This project has the potential of an enabling technology in computational sciences, and the potential to change the way to think of programming and our approach to software construction. Dos Reis will involve students both at the graduate and undergraduate levels, in hands-on, fully integrated research-based classes.

Dos Reis received his Ph.D. in mathematics from the University of Paris VII and École Normale Supérieure de Cachan—France in 2001. He worked as a postdoctoral research associate for INRIA Sophia Antipolis in France and for the Department of Computer Science and Engineering at Texas A&M before joining the faculty at Texas A&M. His research interests include computer algebra, mathematical software, formal verification, programming languages, compiler construction, and generic programming. Dr. Dos Reis is presently a trustee of the Calculemus project; project lead of the OpenAxiom computer algebra system; project lead of the Liz programming system; member of the ISO C++ Standardization committee; and member of the AFNOR (French national body for standardizations) C++ committee.

The NSF established the CAREER program to support junior faculty within the context of their overall career development, combining in a single program the support of research and education of the highest quality in the broadest sense. Through this program, the NSF emphasizes the importance of the early development of academic careers dedicated to simulating the discovery process in which the excitement of research is enhanced by inspired teaching and enthusiastic learning. For more on the NSF and the CAREER program, visit http://www.nsf.gov.