![[2602.19223] Characterizing MARL for Energy Control: A Multi-KPI Benchmark on the CityLearn Environment](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.19223] Characterizing MARL for Energy Control: A Multi-KPI Benchmark on the CityLearn Environment
Summary
This paper presents a comprehensive benchmark for Multi-Agent Reinforcement Learning (MARL) applied to urban energy management using the CityLearn environment, highlighting performance metrics and algorithm comparisons.
Why It Matters
As urban energy systems grow more complex, effective management is crucial for sustainability. This study establishes a benchmark for MARL algorithms, providing insights into their performance and potential for real-world energy control applications, which is vital for developing smart cities.
Key Takeaways
- MARL offers a scalable solution for managing urban energy systems.
- The CityLearn environment provides a realistic simulation for benchmarking.
- Decentralized Training with Decentralized Execution (DTDE) outperforms other methods.
- Novel KPIs enhance the evaluation of energy management algorithms.
- The study reveals robustness to resource removal, indicating resilience.
Computer Science > Artificial Intelligence arXiv:2602.19223 (cs) [Submitted on 22 Feb 2026] Title:Characterizing MARL for Energy Control: A Multi-KPI Benchmark on the CityLearn Environment Authors:Aymen Khouja, Imen Jendoubi, Oumayma Mahjoub, Oussama Mahfoudhi, Claude Formanek, Siddarth Singh, Ruan De Kock View a PDF of the paper titled Characterizing MARL for Energy Control: A Multi-KPI Benchmark on the CityLearn Environment, by Aymen Khouja and 6 other authors View PDF HTML (experimental) Abstract:The optimization of urban energy systems is crucial for the advancement of sustainable and resilient smart cities, which are becoming increasingly complex with multiple decision-making units. To address scalability and coordination concerns, Multi-Agent Reinforcement Learning (MARL) is a promising solution. This paper addresses the imperative need for comprehensive and reliable benchmarking of MARL algorithms on energy management tasks. CityLearn is used as a case study environment because it realistically simulates urban energy systems, incorporates multiple storage systems, and utilizes renewable energy sources. By doing so, our work sets a new standard for evaluation, conducting a comparative study across multiple key performance indicators (KPIs). This approach illuminates the key strengths and weaknesses of various algorithms, moving beyond traditional KPI averaging which often masks critical insights. Our experiments utilize widely accepted baselines such as Proximal Poli...
