Fivos Kalogiannis

I am broadly interested in optimization theory for machine learning, deep learning, and (multi-agent) reinforcement-learning.

I am a CS PhD student at UC San Diego advised by Prof. Mikhail Belkin focusing on optimization for machine learning. Previously, I earned my MS in Computer Science at UC Irvine advised by Prof. Ioannis Panageas where I worked extensively on algorithmic game theory, multi-agent reinforcement learning, and optimization. Before that, I did my undergrad in Electrical and Computer Engineering at the National Technical University of Athens.

I grew up on Lemnos, a remote island of the North Aegean Sea.

news

Sep 18, 2025	Policy gradient can provably bluff! Check my NeurIPS ‘25 paper with Gabriele Farina settling the theoretical problem of convergence of REINFORCE-style algorithms in imperfect information games: Policy Gradient Methods Converge Globally in Imperfect-Information Extensive-Form Games.
May 1, 2025	New ICML ‘25 paper with Manolis Vlatakis and Ian Gemp & Georgios Piliouras tackling multi-agent convex utility RL using the proximal PŁ condition to show convergence in min-max optimization.
Sep 25, 2024	New paper accepted at NeurIPS 2024 with Jingming Yan and Ioannis Panageas! Title: Learning Equilibria in Adversarial Team Markov Games: A Nonconvex-Hidden-Concave Min-Max Optimization Problem (on arxiv, soon).