CV

Education

2011: Graduate from high school with distinction, Athénée Emile Bockstael (Brussels)
2014: Bachelor in Computer Science with the greatest distinction, Université Libre de Bruxelles
2016: Master in Computer Science with the greatest distinction, Artificial Intelligence, Université Libre de Bruxelles (master thesis: Reinforcement Learning in Complex Environments: Evaluating Algorithms on Image Classification, promotor: Dr. Peter Vrancx)
2016- (future): PhD thesis, Improve Sample Efficiency of Reinforcement Learning in Markov Decision Processes by Discovering and Leveraging Structure in the Environment, Vrije Universiteit Brussel (promotors: Prof. Dr. Ann Nowé and Dr. Peter Vrancx)

Notable extra-curricular courses: Image Acquisition and Processing, Natural Language Processing, Digital Circuits Design

The Université Libre de Bruxelles and Vrije Universiteit Brussel are two distinct universities even if they have close ties and regularly collaborate on academic or scientific projects.

Publications

Denis Steckelmacher and Peter Vrancx, An Empirical Comparison of Neural Architectures for Reinforcement Learning in Partially Observable Environments, 27th Benelux Conference on Artificial Intelligence, Universiteit Hasselt, 2015 (link)

Research interests

Artificial Intelligence: Reinforcement Learning (hierarchical, partially observable environments, applied to classification), Neural Networks (recurrent and deep networks), Knowledge Representation, Big Data
High-Performance Computing: Microprocessor Architecture (design, hardware-software co-development, hardware-dependent software optimization), Logic Circuit Design (VHDL, fabrication technologies, experience with Xilinx Vivado), Compiler Design and Optimization Passes, Language Theory

I regularly mix my two main interests (Artificial Intelligence and High-Performance Computing) by developing high-performance, parallelized, distributed and optimized AI-related algorithms. In my master thesis, I describe a distributed Reinforcement Learning algorithm based on Q-Learning.

Projects and Experience

A more complete list of projects I worked on is available on this website.

2015: Peer-reviewed publication, see Publications above
2015: 3-months research internship at the Vrije Universiteit Brussel, COMO lab
2014: Google Summer of Code, Javascript parser, syntax highlighting and code-completion for the KDevelop Itegrated Development Environment
2012-2014: SuperBudget, online price comparator, experience as CIO/CTO in a startup
2013: Google Summer of Code, natural language parser for semantic search queries
2011: QtORM, available on Github, a feature-full C++ Object-Relational Mapper
2011: Google Summer of Code, OpenCL implementation built on top of the Clang compiler
2008: Toy x86_64 operating system with its own file-system, memory manager, thread scheduler and driver model