Hi, I'm Argy

Argyrios Gerogiannis

PhD Candidate in Electrical & Computer Engineering at the University of Illinois at Urbana-Champaign (UIUC)

Download CV
Scroll down

Research

My current interests

My research interests broadly encompass machine learning, with a current focus on reinforcement learning and online learning in non-stationary bandit algorithms. I also have prior experience in imitation learning, inverse reinforcement learning, and generative modeling.

Google Scholar

Reinforcement Learning

  • I like to design models for partially observable environments, algorithms for time-varying settings and, particularly, approaches that learn from expert behavior.

Bandit Algorithms

  • I focus on finding solutions for nonstationary models in parametric and non-parametric settings.

Machine Learning

  • I am interested in generative modeling and its intersection with reinforcement learning.

Publications

Most recent work
Track-MDP: RL for Target Tracking

Track-MDP: Reinforcement Learning for Target Tracking with Controlled Sensing ICASSP, 2025

Adarsh M. Subramaniam, Argyrios Gerogiannis, James Z. Hare, Venugopal V. Veeravalli

Paper
Detection Is All You Need

Detection Is All You Need: A Feasible Optimal Prior-Free Black-Box Approach For Piecewise Stationary Bandits (Under Review)

Argyrios Gerogiannis, Yu-Han Huang, Subhonmesh Bose, Venugopal V. Veeravalli

Paper
Change Detection Based Procedures

Change Detection-Based Procedures for Piecewise Stationary MABs: A Modular Approach (Under Review)

Yu-Han Huang, Argyrios Gerogiannis, Subhonmesh Bose, Venugopal V. Veeravalli

Paper
Is Prior-Free Black-Box Non-Stationary RL Feasible?

Is Prior-Free Black-Box Non-Stationary Reinforcement Learning Feasible? AISTATS, 2025

Argyrios Gerogiannis, Yu-Han Huang, Venugopal V. Veeravalli

Paper
Track-MDP: RL for Target Tracking

Track-MDP: Reinforcement Learning for Target Tracking with Controlled Sensing ICASSP, 2025

Adarsh M. Subramaniam, Argyrios Gerogiannis, James Z. Hare, Venugopal V. Veeravalli

Paper
Detection Is All You Need

Detection Is All You Need: A Feasible Optimal Prior-Free Black-Box Approach For Piecewise Stationary Bandits (Under Review)

Argyrios Gerogiannis, Yu-Han Huang, Subhonmesh Bose, Venugopal V. Veeravalli

Paper

About Me

A brief introduction

I am a PhD candidate in the Electrical and Computer Engineering department at the University of Illinois Urbana-Champaign (UIUC), where I am very fortunate to be advised by Prof. Venugopal Veeravalli. Before that, I graduated with a Diploma (BSc+MEng) in Electrical and Computer Engineering from the University of Patras, at which I had the wonderful opportunity to work with Prof. George Moustakides. For more information, feel free to see my CV.

If you enjoy linguistics (or are simply confused), my full legal first name is Argyrios. I mostly go by Argyris, especially when I visit Greece, due to the weird nature of male Greek names! The anglicized version of my name is Argy, pronounced "R-G", hence the name on this site. However, I am also privileged to have a secret name. I go by all four of my names, but I believe there is a clear winner.

Contact Me

Get in touch

Email

Email me at: argyrger@gmail.com or ag91@illinois.edu

Location

My office is located at:
University of Illinois, 312 CSL 1308 West Main Street, Urbana, IL 61801