Farid Alizadeh

Farid Alizadeh is a Professor at Rutgers School of Business, Department of Management Science and Information Systems. He received his Ph.D. in computer science from the University of Minnesota in 1991 and spent two years as a Postdoctoral associate at the University of California-Berkeley. He is one of the originators of the field of semidefinite programming, which has found numerous applications in areas as broad as quantitative finance, statistical learning theory, computer science, and engineering. He is the recipient of the INFORMS Optimization Society 2014 Farkas Prize. Currently, he is working on statistical learning theory problems where data are imprecise and presented as a probability distribution.

Title of talk: Using Probabilistic Data in Learning and its Consequences

Majority of modern optimization methods, from stochastic gradient descent to ”zero-th order” methods use some kind of approximate first order information. We will overview different methods of obtaining this information including simple stochastic gradient via sampling, robust gradient estimation in adversarial settings, traditional and randomized finite difference methods and more. We will discuss what key properties of these inexact, stochastic first order oracles are useful for convergence analysis of optimization methods that use them.