Enhancing Convergence of Decentralized Gradient Tracking under the KL Property
- URL: http://arxiv.org/abs/2412.09556v1
- Date: Thu, 12 Dec 2024 18:44:36 GMT
- Title: Enhancing Convergence of Decentralized Gradient Tracking under the KL Property
- Authors: Xiaokai Chen, Tianyu Cao, Gesualdo Scutari,
- Abstract summary: We establish convergence of the same type for the notorious decentralized gradient-tracking-based algorithm SONATA.
This matches the convergence behavior of centralized-gradient algorithms except when.
thetain (1/2,1)$, sub rate is certified; and.
textbf(iii)$ when $thetain (1/2,1)$, sub rate is certified; and.
textbf(iii)$ when $thetain (1/2,1)$, sub rate is certified.
- Score: 10.925931212031692
- License:
- Abstract: We study decentralized multiagent optimization over networks, modeled as undirected graphs. The optimization problem consists of minimizing a nonconvex smooth function plus a convex extended-value function, which enforces constraints or extra structure on the solution (e.g., sparsity, low-rank). We further assume that the objective function satisfies the Kurdyka-{\L}ojasiewicz (KL) property, with given exponent $\theta\in [0,1)$. The KL property is satisfied by several (nonconvex) functions of practical interest, e.g., arising from machine learning applications; in the centralized setting, it permits to achieve strong convergence guarantees. Here we establish convergence of the same type for the notorious decentralized gradient-tracking-based algorithm SONATA. Specifically, $\textbf{(i)}$ when $\theta\in (0,1/2]$, the sequence generated by SONATA converges to a stationary solution of the problem at R-linear rate;$ \textbf{(ii)} $when $\theta\in (1/2,1)$, sublinear rate is certified; and finally $\textbf{(iii)}$ when $\theta=0$, the iterates will either converge in a finite number of steps or converges at R-linear rate. This matches the convergence behavior of centralized proximal-gradient algorithms except when $\theta=0$. Numerical results validate our theoretical findings.
Related papers
- Two-Timescale Gradient Descent Ascent Algorithms for Nonconvex Minimax Optimization [77.3396841985172]
We provide a unified analysis of two-timescale gradient ascent (TTGDA) for solving structured non minimax optimization problems.
Our contribution is to design TTGDA algorithms are effective beyond the setting.
arXiv Detail & Related papers (2024-08-21T20:14:54Z) - Optimal and Efficient Algorithms for Decentralized Online Convex Optimization [51.00357162913229]
Decentralized online convex optimization (D-OCO) is designed to minimize a sequence of global loss functions using only local computations and communications.
We develop a novel D-OCO algorithm that can reduce the regret bounds for convex and strongly convex functions to $tildeO(nrho-1/4sqrtT)$ and $tildeO(nrho-1/2log T)$.
Our analysis reveals that the projection-free variant can achieve $O(nT3/4)$ and $O(n
arXiv Detail & Related papers (2024-02-14T13:44:16Z) - Convergence of a Normal Map-based Prox-SGD Method under the KL
Inequality [0.0]
We present a novel map-based algorithm ($mathsfnorMtext-mathsfSGD$) for $symbol$k$ convergence problems.
arXiv Detail & Related papers (2023-05-10T01:12:11Z) - Beyond Uniform Smoothness: A Stopped Analysis of Adaptive SGD [38.221784575853796]
This work considers the problem of finding first-order stationary point of a non atau function with potentially constant smoothness using a gradient.
We develop a technique that allows us to prove $mathcalO(fracmathrmpolylog(T)sigmatT)$ convergence rates without assuming uniform bounds on the noise.
arXiv Detail & Related papers (2023-02-13T18:13:36Z) - ReSQueing Parallel and Private Stochastic Convex Optimization [59.53297063174519]
We introduce a new tool for BFG convex optimization (SCO): a Reweighted Query (ReSQue) estimator for the gradient of a function convolved with a (Gaussian) probability density.
We develop algorithms achieving state-of-the-art complexities for SCO in parallel and private settings.
arXiv Detail & Related papers (2023-01-01T18:51:29Z) - Nonsmooth Nonconvex-Nonconcave Minimax Optimization: Primal-Dual Balancing and Iteration Complexity Analysis [23.80683445944524]
We introduce a novel analysis for PLDA, the key are our newly developed nonsmooth and dual error iterations.
When $thetain [0,12]$, PLDA achieves the optimal $mathcalO()$ under mild assumptions.
arXiv Detail & Related papers (2022-09-22T07:12:48Z) - A first-order primal-dual method with adaptivity to local smoothness [64.62056765216386]
We consider the problem of finding a saddle point for the convex-concave objective $min_x max_y f(x) + langle Ax, yrangle - g*(y)$, where $f$ is a convex function with locally Lipschitz gradient and $g$ is convex and possibly non-smooth.
We propose an adaptive version of the Condat-Vu algorithm, which alternates between primal gradient steps and dual steps.
arXiv Detail & Related papers (2021-10-28T14:19:30Z) - Decentralized Riemannian Gradient Descent on the Stiefel Manifold [39.750623187256735]
We consider a distributed non-sensian optimization where a network of agents aims at minimizing a global function over the Stiefel manifold.
To have exact with constant use, we also propose a decentralized gradient (DRA) for the Stiefel manifold.
arXiv Detail & Related papers (2021-02-14T07:30:23Z) - Proximal Gradient Descent-Ascent: Variable Convergence under K{\L}
Geometry [49.65455534654459]
The finite descent-ascent parameters (GDA) has been widely applied to solve minimax optimization problems.
This paper fills such a gap by studying the convergence of the KL-Lized geometry.
arXiv Detail & Related papers (2021-02-09T05:35:53Z) - S-ADDOPT: Decentralized stochastic first-order optimization over
directed graphs [16.96562173221624]
Decentralized convex optimization is proposed to minimize a sum of smooth and strongly cost functions when the functions are distributed over a directed network nodes.
In particular, we propose thetextbftextttS-ADDOPT algorithm that assumes a first-order oracle at each node.
For decaying step-sizes$mathcalO (1/k)$, we show thattextbfttS-ADDOPT reaches the exact solution subly at$mathcalO (1/k)$ and its convergence is networkally-independent
arXiv Detail & Related papers (2020-05-15T21:14:22Z) - Complexity of Finding Stationary Points of Nonsmooth Nonconvex Functions [84.49087114959872]
We provide the first non-asymptotic analysis for finding stationary points of nonsmooth, nonsmooth functions.
In particular, we study Hadamard semi-differentiable functions, perhaps the largest class of nonsmooth functions.
arXiv Detail & Related papers (2020-02-10T23:23:04Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.