Nathan Friedman

Prof. Friedman's research interests are in the general area of the design and analysis of algorithms and computational complexity. Currently, he is interested in parallel algorithms for non-numeric and semi-numeric problems using models of computation motivated by practical design considerations. A particular problem of interest involves algorithms for sorting and ranking using tournaments which can be implemented by sorting nets. He has also been investigating various string recognition problems.

Many results in concerning the complexity of semi-numeric problems use concepts and results that assume one is working with real numbers. In fact, such numbers do not exist on real computers. Prof. Friedman is interested in the relationship between these results and results which can be established using models of computation which reflect the finitary nature of data represented on actual computers. The orbits of chaotic transformations have many applications in modeling complex phenomena and are often computed numerically in simulating such phenomena. He has been studying relationships between the theoretical and the computer-generated orbits of such transformations to attempt to justify rigorously the use of the computer orbits.