AmirMahdi Ahmadinejad

AmirMahdi Ahmadinejad is a fourth year PhD candidate in the Department of Management Science & Engineering at Stanford University. AmirMahdi is also a member of CS Theory Group at Stanford. He is currently working with Amin Saberi and Aaron Sidford on designing fast iterative methods for a set of fundamental problems in the areas of linear algebra, machine learning, and graph theory. Over the last year, AmirMahdi worked on a classic and important problem in linear algebra, which has numerous applications in a variety of other fields such as probability theory, the theory of dynamical systems, economics, social networks, and demography. This problem dates back to the distinguished Perron-Frobenius theorem of O. Perron (1907) and G. Frobenius (1912), which states that a positive real square matrix has a unique largest eigenvalue, and the corresponding eigenvectors can be chosen to be component-wise positive. Although many methods have been proposed to compute those quantities since then, and there are many works which use these quantities, we hypothesized the first nearly linear time algorithm, in work advised by Amin Saberi and Aaron Sidford.