The L1-regularized Gaussian maximum likelihood method is a common approach for sparse precision matrix estimation but one that poses a computational challenge for high-dimensional datasets. I will present a novel method for performant large-scale sparse precision matrix estimation utilizing the block structures and parallelism in the underlying computations for added performance and scalability. Our numerical examples and comparative results with various modern open-source packages reveal that the proposed algorithm provides orders of magnitude faster runtimes and scales to significantly higher-dimensional datasets.
This lecture will be delivered in person in LSK 306 and also streamed via Zoom. If you do not receive IAM seminar announcements by email, please RSVP here to receive the Zoom link.