Parallel and distributed optimization

Abstract: Due to the explosion in size and complexity of modern datasets, both the decentralized collection or storage of these datasets as well as accompanying parallel and distributed solution methods are either necessary or at least highly desirable. In this talk, I will introduce several ways to move from single threaded optimization algorithms to parallel and distributed approaches with examples: direct parallel and distributed implementation, reformulation of the problem, and new algorithms suitable for parallel and distributed platforms. However, there are some new issues in parallel and distributed computing, and I will talk about asynchronous optimization algorithms, which provide ways to deal with the issue of synchronization and reduce the communication time.