Sebastien Origer
Hi there! My name is Seb and I am a 2023 Young Graduate Trainee in Artificial Intelligence and Inverse Problems.
If there's one thing I'd like you to remember about me, it's my passion for sports. If anyone ever doubts the benefits of sports or just needs a pep talk to get running, I am your guy! I have spent a prolonged period playing sports such as football, badminton, beach-volleyball, tennis, table-tennis and gymnastics. Yet, I also enjoy running, padel and surfing, and hopefully I will keep up this trend well into my later years.
I grew up to Luxembourgish/Belgian parents in a quite luxembourgish village. At 19, I realized that problem-solving was not only quite fun, but also a potential path to a fulfilling career. In the following years, I completed both a Bachelor’s and Master’s degree in aerospace engineering at Delft University of Technology, with a specialization in control & simulation of aerospace vehicles.
During my Master’s, I had the wonderful opportunity to join the ACT as an intern, a step that propelled me into the world of research. As an intern I worked on training neural models to represent both the optimal policy (i.e., the optimal thrust direction) and the value function (i.e., the time of flight) for a time optimal, constant acceleration low-thrust rendezvous by using a data augmentation technique called "backward generation of optimal examples", which led to my first peer-reviewed paper (arXiv). By that time, Guidance & Control Networks already had a special place in my heart which allowed me to write my Master’s thesis in collaboration with the ACT on "Guidance & Control Networks for Time-Optimal Quadcopter Flight" (arXiv).
A clear passion for artificial intelligence grew out of my internship and thesis experience, which is why I am so excited to further deepen my knowledge in the coming year as a Young Graduate Trainee. My job consists in applying recent developments in implicit representations with AI, differential intelligence or physically-informed neural networks to new inverse problems. This work has the potential to tackle previously unsolved problems in physics, aid the design of spacecraft and extract more information from the valuable measurements made in space.
Contact: tni.ase@regiro.neitsabes