Bram De Cooman
Research
Stability Analysis and Performance Improvement of Deep Reinforcement Learning Algorithms
In this Ford Alliance project, we try to develop deep reinforcement learning methods that are used to train virtual drivers for highly-automated and autonomous vehicles on highways (dense traffic). These methods should also allow the enforcement of hard safety constraints to ensure a reliable driving agent is learned. Furthermore we will try to analyze and improve the stability and performance of such methods.
Publications
301 Moved Permanently
Contact information
- Office: B00.58
- Address:
ESAT
Kasteelpark Arenberg 10, bus 2446, B-3001 Leuven, Belgium - Email: bram.decooman@kuleuven.be