Researchers Yan Chen, Tao Li, and Xiaofeng Zong have proposed two algorithms for distributed optimization over a graphon - a continuum of nodes. The goal is to minimize a global cost function that's the sum of individual node costs. They use stochastic gradient descent and gradient tracking methods, establishing bounds on the convergence of these algorithms. For connected graphons, they show that all nodes' states converge to a consensus, and if local costs are strongly convex, they reach the global minimum. --- Why it matters: This work is relevant to researchers in distributed optimization and control systems, as it provides new insights into the behavior of stochastic gradient descent and gradient tracking algorithms on large-scale networks. The results have implications for designing efficient and reliable distributed optimization protocols. Source: https://arxiv.org/abs/2407.02765