{"id":"arxiv/2607.23601","url":"https://arxiv.org/abs/2607.23601","tags":["math.OC"],"score":0,"title":"Exact Worst-case Convergence Rates of Distributed Gradient Tracking Methods","domain":"math.OC","source":"arxiv","authors":["Qiuchen Tian","Li Chai","Jinming Xu"],"reviews":0,"subject":"math.OC","summary":"Different from most existing literature in the analysis of distributed optimization algorithms that reports sufficient convergence conditions leading to a conservative convergence rate, this work provides the exact worst-case convergence rates for two typical gradient tracking algorithms, DIGing and AugDGM. By eigen-decomposition, we show that two algorithms share the same average-state dynamics, while they differ from each other in the gradient tracking subsystems, which entirely govern algorithm convergence. Exploiting the diagonal structure of this decomposition, we reduce the stability…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-07-26","reactions":{}}