I am assistant professor with the
. Department of Electrical and Computer Engineering, Michigan State University (MSU)
My research is in the areas of
machine learning, reinforcement learning, optimization, signal processing, sequential decision making, and risk, and their applications in autonomous networked systems, wireless communications, security and privacy, and system trustworthiness.
I also share common interest with my partner, Ioanna, in the area of urban information systems. To learn more about Ioanna’s research, click
Before joining MSU, I was a postdoctoral researcher with the
Department of Electrical and Systems Engineering, University of Pennsylvania, hosted jointly by Profs. George J. Pappas and Alejandro Ribeiro, and affiliated with GRASP, Alelab, and PRECISE. Prior to that, I was a postdoctoral research associate with the Department of Operations Research and Financial Engineering (ORFE), Princeton University, hosted by Prof. Warren B. Powell, and affiliated with CASTLE Labs. I received the PhD degree in Electrical and Computer Engineering from Rutgers University, where I worked with Prof. Athina P. Petropulu.
I am looking for Open Positions: new PhD students and postdocs to work on problems in machine/reinforcement learning and stochastic optimization/control, with applications in the broad area of . wireless autonomy, and with emphasis on robustness and trustworthiness If interested in working with me, send me an email including your CV and a short, casual description of your work and research interests. My CV
Latest News New paper on showing October 28, 2020 noisy linear convergence of stochastic gradient descent for CV@R-based risk-aware learning under Polyak-Łojasiewicz inequalities submitted. Preprint to be available soon. Another paper on October 21, 2020 actor-only deterministic policy gradient via zeroth-order action-space oracles submitted. Our paper titled “Model-Free Learning of Optimal Ergodic Policies in Wireless Systems” was just accepted to the September 24, 2020 IEEE Transactions on Signal Processing (to appear, check out updated preprint). New (and my first) blog article on the August 7, 2020 connections between policy gradient and zeroth-order methods posted in “Blogs and Discussion”; check it out. Our award-winning work on risk-aware MMSE estimation has been July 17, 2020 accepted for oral presentation at the 2020 INFORMS Annual Meeting; read our preprint in “Publications” (also on arXiv). Our paper on risk-aware LQRs has been July 15, 2020 accepted for presentation at CDC 2020; read our preprint (“Publications”, also on arXiv). Revised preprint of our work on July 12, 2020 Zeroth-order Deterministic Policy Gradient (ZDPG) just uploaded; this also contains a more detailed numerical results section. Our work on relay beamforming with predictive relay selection (published in IEEE Access) has been filed as a US provisional patent application (patent pending). June 23, 2020 Preprint of our work on RL via zeroth-order deterministic policy gradient (submitted) now available in “Publications” (will be updated as needed; stay tuned). June 5, 2020 New paper on successive elimination algorithms for (risk-aware) quantile-based multi-armed bandits submitted (preprint available soon; tune-in!) June 2, 2020 Paper on almost-zero duality gaps and related optimality guarantees in model-free resource allocation for wireless was May 30, 2020 accepted in EUSIPCO 2020. Revised version of our paper on non-parametric structure learning on tree graphical models uploaded in “Publications”; check it out. May 18, 2020
This is the exotic Marathonisi island, a satellite of the also exotic island of Zakynthos (my birthplace)