Adrien Bolland

I am a postdoctoral researcher working on deep reinforcement learning and its application to the operation of electrical power systems. I trained as an electrical engineer at the University of Liège, where I completed my Ph.D. under the supervision of Damien Ernst in 2025. My thesis studied how exploration influences policy-gradient algorithms and proposed new approaches to maximum-entropy exploration. Along the way, I worked with industrial and academic partners on energy applications such as microgrid design and control, battery bidding on intraday electricity markets, and renewable energy communities.

As an F.R.S.-FNRS postdoctoral researcher in the group of Louis Wehenkel, I then studied how reinforcement learning can help transmission system operators keep voltages stable as grids integrate more renewable sources and new loads. I now continue this work at École Polytechnique with Paul Mangold, with particular attention to decentralized and multi-agent reinforcement learning.

The animation at the top of this page shows two controllers trained with reinforcement learning, a single agent and a team of agents, acting on an electrical grid to prevent an overvoltage that would otherwise lead to system collapse within a few minutes.

Academic CV (last modified: 8 November 2026)

Adrien Bolland

Selected publications

Full list on Google Scholar.

Presentations & Courses