I am currently a Post-doctoral fellow in Machine Learning and Robotics at the Computational Intelligence Laboratory, Department of Mathematics, University of Patras, in collaboration with Prof. Michael N. Vrahatis. My work focuses on using evolutionary and machine learning methods combined with simulations to develop trial-and-error methods that work on physical robots.
I am also the Leader of the R&D Computer Vision Team at Metargus, a pre-seed funded start up (based in Patras, Greece), building a cutting-edge basketball coaching tool to provide coaches with insights far beyond traditional analytics. My work focuses on combining traditional Computer Vision techniques with modern Machine/Deep Learning algorithms.
I am also currently teaching the under-graduate courses "Introduction to Artificial Intelligence" and "Robotics and Intelligent Agents" at the Computer Engineering & Informatics Department of University of Patras.
I was previously a Post-doctoral fellow in Machine Learning and Robotics at the Computer Technology Institute, Patras in collaboration with Prof. Paul Spirakis and Theodoros Komninos. My work focused on effectively combining reinforcement learning methods with traditional machine learning and game theoretic methods for data-efficient learning.
I was also previously a Post-doctoral fellow in Machine Learning and Robotics at the LASA team at EPFL under the supervision of Prof. Aude Billard. My work focused on effectively combining reinforcement learning methods with traditional robotics control and dynamical system-based learning methods for data-efficient learning. My work was funded by the ERC "SAHR" and the CHIST-ERA "CORSMAL" projects.
I defended my PhD in Robotics and Machine Learning entitled ''Micro-Data Reinforcement Learning for Adaptive Robots'' on 14th December 2018 (University of Lorraine and Inria Nancy, LARSEN Team under the supervision of Dr. Jean-Baptiste Mouret). My work focused on studying and providing abilities to autonomous robots to be able to compensate for unknown situations (such as damages) using machine learning techniques and evolutionary computation. My work was funded by the ERC "ResiBots" project.