Axel Faes
PhD Fellow

An Information Theoretical Approach to EEG Source-Reconstructed Connectivity

Advanced Master's Thesis
Feb. 2018 - Jul. 2018

This thesis takes an information theoretical approach, which concerns model-free, probability based methods such as Conditional Mutual Information, Directed Information, and Directed feature information. - 17/20

Algebraic Subtyping for Algebraic Effects and Handlers

Master's Thesis
Feb. 2018 - Jul. 2018

Extending Algebraic Subtyping to incorperate support for algebraic effects and handlers. Final score - 19/20

Reinforcement Learning Agent in Google Deepmind’s StarCraft II Framework - CSAI

Feb. 2018 - Jul. 2018

Implement several learning algorithms in PySC2

Software Architecture course - Project

Feb. 2017 - Jul. 2017

Project made for the course ’Software Architecture’. The goal was to design a software architecture in UML for a IoT-platform concerning pluggable sensors. The platform allows storage of customer data and the use of third party applications for data analytics. Final score - 18/20

ICAL parser for KU Leuven schedules

Lead Developer
Aug. 2016 - Current

An nodejs application to create an iCalender file for courses at KU Leuven. Allows the creation of a schedule containing courses from different masters and the option to ignore events.

Machine learning techniques for flow-based network intrusion detection systems

Bachelor's Thesis
Feb. 2016 - Jul. 2016

The thesis gives an overview of how machine learning algorithms could be used for intrusion detection using only IP Flows. The system has been used to detect intrusions in Cegeka Hasselt Datacenter network.

Software engineering: Search and Recommendation System

Team Member
Feb. 2016 - Jul. 2016

A search and recommendation system for VoD (Video on Demand) for Androme. The system is currently being used in production in the Nebula project. Both Content-Based Recommendations and Collaborative filtering techniques were implemented. Made in a team of 5 (Pieter Teunen, Luuk Raaijmakers, Brent Berghmans, Axel Faes, Matthijs Kaminski, Wouter Bollaert) utilising Java and the Spring framework. Final score - 15/20

TTUI: Household Survival

Sep. 2015 - Dec. 2015

Project made for the class ‘Technologies and Tools for User Interfaces’.
A tower-defense style game written in Unity utilising Optitrack motion capture. The game combines the virtual world and reality, by allowing users to interact with the virtual world using real-world objects. Made by Brent Berghmans, Axel Faes and Matthijs Kaminski. Final score - 18/20

Cardinal: scripting language

Lead Developer
Jan. 2015 - Sep. 2015

Cardinal is a small, fast, class-based, Object Oriented scripting language written in C. It is built upon the skeleton of an existing scripting language and shows how I can modify and improve existing software, as well as design new components to this software.

New components include a debugger, an embedding API, multiple inheritance and a new module system.

United in Manchester

Team Leader
Jul. 2015 - Aug. 2015

A summer school which focuses on teamwork in cross-cultural and multidisciplinary teams, global product development and entrepreneurship. Our team developed a start-up idea on Food Management/Delivery system. Product pitch took place at the end of the course for feedbacks from professionals. Our team consisted of Axel Faes, Linh Chi Evelyn Phan, Reinaert Van de Cruys and Maria Barouh.

PSOPV: Visual Programming IDE

Feb. 2015 - Jul. 2015

A Visual programming IDE created by Axel Faes & Matthijs Kaminski for a course of Hasselt University. The purpose of the IDE is to create ‘black boxes’ which can send events (signals packed with data) to eachother. We take the idea of using drag-able blocks in a visual IDE and expand on it. Final score - 17/20