{"id":"arxiv/2606.20356","url":"https://arxiv.org/abs/2606.20356","tags":["math.OC","cs.AI","cs.LG","math.PR","stat.ML"],"score":0,"title":"Robust $Q$-learning for mean-field control under Wasserstein uncertainty in common noise","domain":"math.OC","source":"arxiv","authors":["Mathieu Laurière","Ariel Neufeld","Kyunghyun Park"],"reviews":0,"subject":"math.OC","summary":"In this article, we present a robust $Q$-learning algorithm for discrete-time mean-field control problems under Wasserstein uncertainty in the common noise law. The algorithm combines a quantization-and-projection scheme with a Wasserstein dual reformulation on the common-noise space. We establish its convergence together with finite-time iteration bounds for both synchronous and asynchronous learning schemes. Numerical experiments on systemic risk and epidemic models compare the asynchronous implementation with an idealized Bellman iteration, illustrate the robustness-performance tradeoff…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-18","reactions":{}}