Ph.D. (2015) University of California, Los Angeles
Damek analyzes and develops algorithms for solving optimization problems that arise in machine learning and signal processing. These real applications inevitably lead to complex problems that are nonconvex, nonsmooth, nonlinear, and generally large-scale. To solve these problems, Damek constructs algorithms that adapt to modern computer architectures, which lead to parallelized and even decentralized software implementations. After developing these algorithms, Damek formulates and proves theoretical guarantees on their best- and worst-case behavior.