We are seeking motivated students with a passion for AI and energy systems to join the cutting-edge research project ADAPT on electricity market simulations with multi-agent reinforcement learning (RL).

Background:

Electricity markets are undergoing a radical transformation, with renewable energy and decentralized participants (e.g., prosumers, battery storage) creating highly dynamic, multi-agent environments. Multi-agent RL [1] can be applied to analyze emerging market dynamics. The ASSUME toolbox [2] enables electricity market simulations using RL, yet scaling to large scenarios remains a challenge due to computational inefficiency, non-stationarity, and unstable training dynamics.

Objectives:

The thesis will focus on enabling scalable multi-agent RL simulations by:

• Integrating state-of-the-art machine learning methods into the ASSUME framework, such as parameter sharing, curriculum learning, or entropy regularization.
• Evaluating the benefits and limitations of these methods in the context of multi-agent RL for electricity markets.

Requirements:

• Strong Python coding skills and interest in machine learning applications
• Experience with reinforcement learning and electricity markets is a plus

Formalities:

The thesis can be written in German or English. Please apply with a short letter of motivation (max. ½ page), your CV, and a current grade transcript. Work can commence immediately.

Relevant introductory literature:

[1] Gronauer, Sven, and Klaus Diepold. 2022. “Multi-Agent Deep Reinforcement Learning: A Survey.” Artificial Intelligence Review 55 (2): 895–943. https://doi.org/10.1007/s10462-021-09996-w.

[2] Harder, Nick, Kim K. Miskiw, Manish Khanra, et al. 2025. “ASSUME: An agent-based simulation framework for exploring electricity market dynamics with reinforcement learning.” SoftwareX 30: 102176. https://doi.org/10.1016/j.softx.2025.102176.