Armacki, A., Cai, H., & Sayed, A. H. (2026). Decentralized SGD under Heavy-Tailed Noise: Optimal Convergence Rates and the Role of Gradient Clipping. https://omanscience.com/en/articles/decentralized-sgd-under-heavy-tailed-noise-optimal-convergence-rates-and-the-role-of-gradient-clipping