From Gradient Clipping to Normalization for Heavy Tailed SGD

• URL: http://arxiv.org/abs/2410.13849v1
• Date: Thu, 17 Oct 2024 17:59:01 GMT
• Title: From Gradient Clipping to Normalization for Heavy Tailed SGD
• Authors: Florian Hübler, Ilyas Fatkhullin, Niao He,
• Abstract summary: Recent empirical evidence indicates that machine learning applications involve heavy-tailed noise, which challenges the standard assumptions of bounded variance in practice. In this paper, we show that it is possible to achieve tightness of the gradient-dependent noise convergence problem under tailed noise.
• Score: 19.369399536643773
• License:
• Abstract: Recent empirical evidence indicates that many machine learning applications involve heavy-tailed gradient noise, which challenges the standard assumptions of bounded variance in stochastic optimization. Gradient clipping has emerged as a popular tool to handle this heavy-tailed noise, as it achieves good performance in this setting both theoretically and practically. However, our current theoretical understanding of non-convex gradient clipping has three main shortcomings. First, the theory hinges on large, increasing clipping thresholds, which are in stark contrast to the small constant clipping thresholds employed in practice. Second, clipping thresholds require knowledge of problem-dependent parameters to guarantee convergence. Lastly, even with this knowledge, current sampling complexity upper bounds for the method are sub-optimal in nearly all parameters. To address these issues, we study convergence of Normalized SGD (NSGD). First, we establish a parameter-free sample complexity for NSGD of $\mathcal{O}\left(\varepsilon^{-\frac{2p}{p-1}}\right)$ to find an $\varepsilon$-stationary point. Furthermore, we prove tightness of this result, by providing a matching algorithm-specific lower bound. In the setting where all problem parameters are known, we show this complexity is improved to $\mathcal{O}\left(\varepsilon^{-\frac{3p-2}{p-1}}\right)$, matching the previously known lower bound for all first-order methods in all problem dependent parameters. Finally, we establish high-probability convergence of NSGD with a mild logarithmic dependence on the failure probability. Our work complements the studies of gradient clipping under heavy tailed noise improving the sample complexities of existing algorithms and offering an alternative mechanism to achieve high probability convergence.

Related papers

• Convergence Analysis of Adaptive Gradient Methods under Refined Smoothness and Noise Assumptions [18.47705532817026]
  We show that AdaGrad outperforms SGD by a factor of $d$ under certain conditions. Motivated by this, we introduce assumptions on the smoothness structure of the objective and the gradient variance.
  arXiv  Detail & Related papers  (2024-06-07T02:55:57Z)
• Breaking the Heavy-Tailed Noise Barrier in Stochastic Optimization Problems [56.86067111855056]
  We consider clipped optimization problems with heavy-tailed noise with structured density. We show that it is possible to get faster rates of convergence than $mathcalO(K-(alpha - 1)/alpha)$, when the gradients have finite moments of order. We prove that the resulting estimates have negligible bias and controllable variance.
  arXiv  Detail & Related papers  (2023-11-07T17:39:17Z)
• Parameter-Agnostic Optimization under Relaxed Smoothness [25.608968462899316]
  We show that Normalized Gradient Descent with Momentum (NSGD-M) can achieve a rate-optimal complexity without prior knowledge of any problem parameter. In deterministic settings, the exponential factor can be neutralized by employing Gradient Descent with a Backtracking Line Search.
  arXiv  Detail & Related papers  (2023-11-06T16:39:53Z)
• SGD with AdaGrad Stepsizes: Full Adaptivity with High Probability to Unknown Parameters, Unbounded Gradients and Affine Variance [33.593203156666746]
  We show that AdaGrad stepsizes a popular adaptive (self-tuning) method for first-order optimization. In both the low-noise and high-regimes we find sharp rates of convergence in both the low-noise and high-regimes.
  arXiv  Detail & Related papers  (2023-02-17T09:46:08Z)
• High-Probability Bounds for Stochastic Optimization and Variational Inequalities: the Case of Unbounded Variance [59.211456992422136]
  We propose algorithms with high-probability convergence results under less restrictive assumptions. These results justify the usage of the considered methods for solving problems that do not fit standard functional classes in optimization.
  arXiv  Detail & Related papers  (2023-02-02T10:37:23Z)
• Clipped Stochastic Methods for Variational Inequalities with Heavy-Tailed Noise [64.85879194013407]
  We prove the first high-probability results with logarithmic dependence on the confidence level for methods for solving monotone and structured non-monotone VIPs. Our results match the best-known ones in the light-tails case and are novel for structured non-monotone problems. In addition, we numerically validate that the gradient noise of many practical formulations is heavy-tailed and show that clipping improves the performance of SEG/SGDA.
  arXiv  Detail & Related papers  (2022-06-02T15:21:55Z)
• High-probability Bounds for Non-Convex Stochastic Optimization with Heavy Tails [55.561406656549686]
  We consider non- Hilbert optimization using first-order algorithms for which the gradient estimates may have tails. We show that a combination of gradient, momentum, and normalized gradient descent convergence to critical points in high-probability with best-known iteration for smooth losses.
  arXiv  Detail & Related papers  (2021-06-28T00:17:01Z)
• High Probability Complexity Bounds for Non-Smooth Stochastic Optimization with Heavy-Tailed Noise [51.31435087414348]
  It is essential to theoretically guarantee that algorithms provide small objective residual with high probability. Existing methods for non-smooth convex optimization have complexity bounds with dependence on confidence level. We propose novel stepsize rules for two methods with gradient clipping.
  arXiv  Detail & Related papers  (2021-06-10T17:54:21Z)
• 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)

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.