The Role of Gradient Modification in Heavy-Tailed Nonconvex Stochastic Min-Max Optimization

• URL: http://arxiv.org/abs/2609.06064v1
• Date: Sat, 05 Sep 2026 12:49:28 GMT
• Title: The Role of Gradient Modification in Heavy-Tailed Nonconvex Stochastic Min-Max Optimization
• Abstract summary: min-max optimization has attracted increasing attention due to its applications in modern machine learning.<n>We provide a comprehensive theoretical of gradient descent (SGDA)<n>We develop new clipping-free algorithms, i.e., Stoc-GDAM and Stoc-TRmax, and they both can achieve the optimal dependence on the target accuracy without using gradient clipping.
• Score: 50.59470683896735
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Stochastic min-max optimization has attracted increasing attention due to its applications in modern machine learning, while existing theoretical studies mainly rely on the bounded variance assumption for stochastic gradients. Under heavy-tailed noise, where stochastic gradients only possess a finite $p$-th moment for $p\in(1,2]$, gradient clipping or normalization is commonly believed to be necessary to guarantee convergence. In this work, we revisit stochastic min-max optimization under heavy-tailed noise and provide a comprehensive theoretical study of stochastic gradient descent ascent (SGDA). We first show that vanilla SGDA, without any modification to its update rule, can converge under heavy-tailed noise in both nonconvex-strongly-concave (NC-SC) and nonconvex-concave (NC-C) settings, establishing the first convergence guarantees for SGDA in these regimes. Beyond unregularized problems, we further investigate regularized stochastic min-max optimization, where directly incorporating gradient normalization into proximal updates is nontrivial due to the incompatibility between normalization and proximal structures. We overcome this difficulty by developing new clipping-free algorithms, i.e., Stoc-TRGDAM and Stoc-TRGDmax, and they both can achieve the optimal dependence on the target accuracy without using gradient clipping.

Related papers

• In-Expectation Convergence of Stochastic Gradient Methods under Heavy-Tailed Noise [3.8357180714081327]
  Many methods are believed not to converge when the noise in gradients has only a finite $p$-th moment.<n> Gradient Descent ($textSG$) can surprisingly converge in expectation for convex problems with bounded domains.
  arXiv  Detail & Related papers  (2026-05-30T04:27:47Z)
• Gradient Normalization Provably Benefits Nonconvex SGD under Heavy-Tailed Noise [60.92029979853314]
  We investigate the roles of gradient normalization and clipping in ensuring the convergence of Gradient Descent (SGD) under heavy-tailed noise. Our work provides the first theoretical evidence demonstrating the benefits of gradient normalization in SGD under heavy-tailed noise. We introduce an accelerated SGD variant incorporating gradient normalization and clipping, further enhancing convergence rates under heavy-tailed noise.
  arXiv  Detail & Related papers  (2024-10-21T22:40:42Z)
• Provable Complexity Improvement of AdaGrad over SGD: Upper and Lower Bounds in Stochastic Non-Convex Optimization [18.47705532817026]
  Adaptive gradient methods are among the most successful neural network training algorithms.<n>These methods are known to achieve better dimensional dependence than over convex SGD-normarity.<n>In this paper we introduce new assumptions on the smoothness of the structure and the gradient noise variance.
  arXiv  Detail & Related papers  (2024-06-07T02:55:57Z)
• The Power of Adaptivity in SGD: Self-Tuning Step Sizes with Unbounded Gradients and Affine Variance [46.15915820243487]
  We show that AdaGrad-Norm exhibits an order optimal convergence of $mathcalOleft. We show that AdaGrad-Norm exhibits an order optimal convergence of $mathcalOleft.
  arXiv  Detail & Related papers  (2022-02-11T17:37:54Z)
• On the Double Descent of Random Features Models Trained with SGD [78.0918823643911]
  We study properties of random features (RF) regression in high dimensions optimized by gradient descent (SGD) We derive precise non-asymptotic error bounds of RF regression under both constant and adaptive step-size SGD setting. We observe the double descent phenomenon both theoretically and empirically.
  arXiv  Detail & Related papers  (2021-10-13T17:47:39Z)
• Differentiable Annealed Importance Sampling and the Perils of Gradient Noise [68.44523807580438]
  Annealed importance sampling (AIS) and related algorithms are highly effective tools for marginal likelihood estimation. Differentiability is a desirable property as it would admit the possibility of optimizing marginal likelihood as an objective. We propose a differentiable algorithm by abandoning Metropolis-Hastings steps, which further unlocks mini-batch computation.
  arXiv  Detail & Related papers  (2021-07-21T17:10:14Z)
• Stochastic Optimization with Heavy-Tailed Noise via Accelerated Gradient Clipping [69.9674326582747]
  We propose a new accelerated first-order method called clipped-SSTM for smooth convex optimization with heavy-tailed distributed noise in gradients. We prove new complexity that outperform state-of-the-art results in this case. We derive the first non-trivial high-probability complexity bounds for SGD with clipping without light-tails assumption on the noise.
  arXiv  Detail & Related papers  (2020-05-21T17:05:27Z)
• An Optimal Multistage Stochastic Gradient Method for Minimax Problems [8.615625517708324]
  We study the minimax optimization problem in the smooth and strongly convex-strongly concave setting. We first analyze the Gradient Descent Ascent (GDA) method with constant stepsize. We propose a multistage variant of GDA that runs in multiple stages with a particular learning rate decay schedule.
  arXiv  Detail & Related papers  (2020-02-13T18:01:18Z)
• Optimal Epoch Stochastic Gradient Descent Ascent Methods for Min-Max Optimization [61.66927778694731]
  Epoch gradient descent method (a.k.a. Epoch-GD) proposed by Hazan and Kale (2011) was deemed a breakthrough for strongly convex minimization. Our result is the first one that shows Epoch-GDA can achieve the optimal rate of $O (1/T)$ for the duality gap of SCSC min-max problems.
  arXiv  Detail & Related papers  (2020-02-13T02:16:57Z)
• Towards Better Understanding of Adaptive Gradient Algorithms in Generative Adversarial Nets [71.05306664267832]
  Adaptive algorithms perform gradient updates using the history of gradients and are ubiquitous in training deep neural networks. In this paper we analyze a variant of OptimisticOA algorithm for nonconcave minmax problems. Our experiments show that adaptive GAN non-adaptive gradient algorithms can be observed empirically.
  arXiv  Detail & Related papers  (2019-12-26T22:10:10Z)

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.