In this talk, we develop distributed Alternating Direction Method of Multipliers (ADMM) based methods for solving general separable convex problems in large-scale systems, which can be applied to the LASSO and many other machine learning problems. We present both synchronous and asynchronous versions of the algorithm and show that both achieves the best known rate of convergence. We also study the relationship between convergence speed and underlying network topology.

Bio: Ermin Wei is currently an Assistant Professor within the EE Academic Program at the EECS Department of Northwestern University. She completed her PhD studies in Electrical Engineering and Computer Science at MIT in 2014, advised by Professor Asu Ozdaglar, where she also obtained her M.S.. She received her undergraduate triple degree in Computer Engineering, Finance and Mathematics with a minor in German, from University of Maryland, College Park. Wei has received many awards, including the Graduate Women of Excellence Award, second place prize in Ernst A. Guillemen Thesis Award and Alpha Lambda Delta National Academic Honor Society Betty Jo Budson Fellowship. Wei’s research interests include distributed optimization methods, convex optimization and analysis, smart grid and energy networks and market economic analysis.