Alexandre Proutiere 2026 ACM SIGMETRICS Achievement Award
ACM SIGMETRICS is pleased to announce the selection of Professor Alexandre Proutiere of KTH, the Royal Institute of Technology, as the recipient of the 2026 ACM SIGMETRICS Achievement Award in recognition of his outstanding contributions to the analysis, evaluation, and design of algorithms for large-scale computer and communication systems.
Prof. Proutière works at the Decision and Control Systems department in the School of Electrical Engineering and Computer Science at KTH Royal Institute of Technology since 2011, and at Chalmers University of Technology since 2025. Before that, he was a permanent researcher at Microsoft Research, Cambridge (UK). His research interests include communication networks and machine learning. His work has led to significant advancements in these fields, particularly in the areas of performance evaluation and design of communication systems, bandit optimization, and reinforcement learning.
Proutière’s work has been influential across both networking and machine learning communities. His contributions have been recognized by numerous awards, including two Best Paper Awards at ACM SIGMETRICS and the 2009 ACM SIGMETRICS Rising Star Award for his contributions to the analysis and design of distributed control mechanisms in networks. His early contributions to communication systems include the development of a comprehensive flow-level theory of the Internet, providing explicit formulas for user-perceived performance and introducing together with Thomas Bonald the notion of balanced fairness. He also made fundamental contributions to the analysis of decentralized multiple access protocols, solving long-standing open problems on their stability using mean-field techniques.
More recently, Prof. Proutière has made major contributions to machine learning, particularly in bandit optimization and reinforcement learning. His work has established instance-specific optimality results and led to the design of algorithms achieving fundamental performance limits across a wide range of models. These results have had significant practical impact, with applications in communication systems, recommender systems, and cellular network optimization.
His recent research further bridges reinforcement learning and control theory, providing finite-time guarantees for classical control problems and developing algorithms with optimal regret. In parallel, he has contributed to unsupervised learning and low-rank matrix estimation, designing efficient algorithms with strong statistical guarantees for uncovering structure in complex data.
Additional information is available at https://people.kth.se/~alepro/
