We study the consensus decentralized optimization problem where the objective function is the average of $n$ agents private non-convex cost functions; moreover, the agents can only communicate to their neighbors on a given network topology. The stochastic learning setting is considered in this paper where each agent can only access a noisy estimate of its gradient. Many decentralized methods can solve such problem including EXTRA, Exact-Diffusion/D$^2$, and gradient-tracking. Unlike the famed DSGD algorithm, these methods have been shown to be robust to the heterogeneity across the local cost functions. However, the established convergence rates for these methods indicate that their sensitivity to the network topology is worse than DSGD. Such theoretical results imply that these methods can perform much worse than DSGD over sparse networks, which, however, contradicts empirical experiments where DSGD is observed to be more sensitive to the network topology. In this work, we study a general stochastic unified decentralized algorithm (SUDA) that includes the above methods as special cases. We establish the convergence of SUDA under both non-convex and the Polyak-Lojasiewicz condition settings. Our results provide improved network topology dependent bounds for these methods (such as Exact-Diffusion/D$^2$ and gradient-tracking) compared with existing literature. Moreover, our results show that these methods are often less sensitive to the network topology compared to DSGD, which agrees with numerical experiments.
翻译:我们研究的是共识分散化优化问题,在这种问题上,目标功能是平均一美元代理商的私人非碳化成本功能;此外,代理商只能通过特定的网络地形学向邻居传达;本文考虑的是每个代理商只能获取对其梯度的噪音估计值的随机学习环境。许多分散化的方法可以解决这类问题,包括Extra、Exact-Dubil/D$2美元和梯度跟踪。与著名的DSGD算法不同,这些方法已证明能够有力地适应当地成本函数的异质性。然而,这些方法的既定趋同率表明,他们对网络地形学的敏感性比DSGD差得多。这些理论结果表明,这些方法比DSGD的偏差得多,然而,这与实验性实验方法相矛盾,DSGD被认为对网络学更为敏感。 我们研究的是,一般的杂质统一分散化算法(SUDA)将上述方法列为特殊案例。我们根据SUDA对网络的敏感度和顶层-Sloja-lical-Lia 的计算结果,这些比Slimia-liumal-I-Lisal-Lislation-Lislation-Lislex-Lislational 提供这些结果,这些方法与我们的最高-Lisl-Lisl-I-I-Lisl-I-I-Lisl-Lisl-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-I-