Title: Bandit change-point detection and its application

Abstract: We investigate the problem of bandit change-point problem when monitoring high-dimensional streaming data in resources constrained environments, where one has limited capacity in data acquisition, transmission or processing, and needs decide how to smartly observe which local components or features of high-dimensional streaming data at each and every time so as to detect changes quickly. In the first part of this talk, we provide an overview of the classical sequential change-point detection problem for one-dimensional data. In the second part of the talk, we present our latest research on efficient scalable schemes for online monitoring high-dimensional data, as well as the corresponding multi-armed bandit versions. Asymptotic analysis, numerical simulations, and case studies will be presented to demonstrate the usefulness of our proposed approach.