PhD scholarship in Multi-agent constrained reinforcement learning

This project addresses the critical technical challenges involved in transitioning from single-agent and unconstrained learning to multi-agent and constrained learning. In a natural manner, due to the autonomy component, this learning process will take the form of reinforcement learning.

The successful candidate will be involved in research related to connected autonomy. More specifically, the research project will consist of the development of new reinforcement learning algorithms designed to work in multi-agent scenarios and under operational requirements. Potential experimental validations of the developed algorithms are also considered.

Ideal candidates should hold a M.Sc. degree in Electrical Engineering, Computer Science or any related discipline. Be fluent in English and possess a solid academic record with strong mathematical and problem-solving abilities. Experience in Python with machine learning libraries (PyTorch) and/or robotic libraries (ROS) is highly valued.