High dimensional analysis reveals conservative sharpening and a stochastic edge of stability

• URL: http://arxiv.org/abs/2404.19261v1
• Date: Tue, 30 Apr 2024 04:54:15 GMT
• Title: High dimensional analysis reveals conservative sharpening and a stochastic edge of stability
• Abstract summary: We show that the dynamics of the large eigenvalues of the training loss Hessian have some remarkably robust features across models and in the full batch regime. There is often an early period of progressive sharpening where the large eigenvalues increase, followed by stabilization at a predictable value known as the edge of stability.
• Score: 21.12433806766051
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent empirical and theoretical work has shown that the dynamics of the large eigenvalues of the training loss Hessian have some remarkably robust features across models and datasets in the full batch regime. There is often an early period of progressive sharpening where the large eigenvalues increase, followed by stabilization at a predictable value known as the edge of stability. Previous work showed that in the stochastic setting, the eigenvalues increase more slowly - a phenomenon we call conservative sharpening. We provide a theoretical analysis of a simple high-dimensional model which shows the origin of this slowdown. We also show that there is an alternative stochastic edge of stability which arises at small batch size that is sensitive to the trace of the Neural Tangent Kernel rather than the large Hessian eigenvalues. We conduct an experimental study which highlights the qualitative differences from the full batch phenomenology, and suggests that controlling the stochastic edge of stability can help optimization.

Related papers

• Information Geometric Self-Organization at the Edge of Stability in High-Capacity Kernel Associative Memories [0.0]
  We show that optimal high-capacity memory representations are not formed in flat minima, but are sculpted at the highly curved boundaries of geometric singularities.<n>These findings suggest that optimal, high-capacity memory representations are not formed in flat minima, but are dynamically sculpted at the highly curved boundaries of geometric singularities.
  arXiv  Detail & Related papers  (2026-09-15T08:27:55Z)
• The Road Taken: The Role of Optimizers at the Edge of Stability [53.81268610971847]
  edge of stability refers to a phenomenon in deep learning where the Hessian eigenvalues of the loss remain stable above a threshold that the classical descent predicts to be unstable.<n>Previous works formulate the edge of stability with respect to the maximum Hessian eigenvalue and the learning rate.<n>We observe that many first-order methods, including gradient descent, significantly violate the stability bound predicted by these theories by a factor as large as $times 21.1$.
  arXiv  Detail & Related papers  (2026-08-19T01:03:34Z)
• Eigenanalysis framework for autoregressive neural emulators of multi-scale chaotic dynamics [37.134314242843914]
  We develop an eigenanalysis framework that reveals the dynamical origin of error growth.<n>We show how inference-time error growth is governed by its spectral radius.<n>Our results establish a theoretical foundation for the design and evaluation of neural emulators of chaotic multi-scale dynamics.
  arXiv  Detail & Related papers  (2026-08-17T04:30:03Z)
• SGD at the Edge of Stability: Stochastic Stabilization with Large Learning Rates [2.104588729923885]
  Modern deep learning has been shown to operate at the edge of stability.<n>We provide convergence guarantees for Gradient Descent (SGD) applied to the multiclass cross-entropy loss.<n> Experiments validate our theoretical findings and illustrate the benefits of SGD in the large-stepsize regime.
  arXiv  Detail & Related papers  (2026-06-29T21:32:58Z)
• Taming Curvature: Architecture Warm-Up for Stable Transformer Training [52.6838951211596]
  We introduce a fast online estimator of the largest (preconditioned) Hessian eigenvalue (i.e., curvature) based on a warm-started variant for power with Hessian-vector products.<n>We find that training instabilities coincide with surges in preconditioned curvature and that curvature grows with depth.<n> Experiments on large Transformers validate that our approach enables efficient curvature tracking and reduces instabilities.
  arXiv  Detail & Related papers  (2026-06-15T14:16:56Z)
• Detecting Deepfakes via Hamiltonian Dynamics [58.881686693914446]
  We propose a new perspective on deepfake detection: moving from static pattern recognition to dynamical stability analysis.<n>Our approach is motivated by physics-inspired priors.<n>Hamiltonian Action Anomaly Detection (HAAD) outperforms evaluated state-of-the-art baselines.
  arXiv  Detail & Related papers  (2026-05-06T01:55:38Z)
• Zeroth-Order Optimization at the Edge of Stability [59.54782674222443]
  We provide an explicit step size condition that captures the (mean-square) linear stability of a family of Zeroth-order (ZO) methods.<n>Our results highlight an implicit regularization effect specific to ZO methods, where large step sizes primarily regularize the Hessian trace.
  arXiv  Detail & Related papers  (2026-04-16T06:23:18Z)
• On Forgetting and Stability of Score-based Generative models [6.259598237089842]
  Understanding the stability and long-time behavior of generative models is a fundamental problem in modern machine learning.<n>This paper provides quantitative bounds on the sampling error of score-based generative models by leveraging stability and forgetting properties of the Markov chain associated with the reverse-time dynamics.
  arXiv  Detail & Related papers  (2026-01-29T15:37:50Z)
• Stability as a Liability:Systematic Breakdown of Linguistic Structure in LLMs [5.96875296117642]
  We show that stable parameter trajectories lead stationary solutions to minimize the forward KL divergence to the empirical distribution.<n>We empirically validate this effect using a controlled feedback-based training framework.<n>It indicates that optimization stability and generative expressivity are not inherently aligned, and that stability alone is an insufficient indicator of generative quality.
  arXiv  Detail & Related papers  (2026-01-26T15:34:50Z)
• Can Stability be Detrimental? Better Generalization through Gradient Descent Instabilities [14.741581246137404]
  We show that instabilities induced by large learning rates move model parameters toward flatter regions of the loss landscape. We find these lead to excellent generalization performance on modern benchmark datasets.
  arXiv  Detail & Related papers  (2024-12-23T14:32:53Z)
• Regulating Model Reliance on Non-Robust Features by Smoothing Input Marginal Density [93.32594873253534]
  Trustworthy machine learning requires meticulous regulation of model reliance on non-robust features. We propose a framework to delineate and regulate such features by attributing model predictions to the input.
  arXiv  Detail & Related papers  (2024-07-05T09:16:56Z)
• Stable Update of Regression Trees [0.0]
  We focus on the stability of an inherently explainable machine learning method, namely regression trees. We propose a regularization method, where data points are weighted based on the uncertainty in the initial model. Results show that the proposed update method improves stability while achieving similar or better predictive performance.
  arXiv  Detail & Related papers  (2024-02-21T09:41:56Z)
• A Precise Characterization of SGD Stability Using Loss Surface Geometry [8.942671556572073]
  Descent Gradient (SGD) stands as a cornerstone optimization algorithm with proven real-world empirical successes but relatively limited theoretical understanding. Recent research has illuminated a key factor contributing to its practical efficacy: the implicit regularization it instigates.
  arXiv  Detail & Related papers  (2024-01-22T19:46:30Z)
• Universal Sharpness Dynamics in Neural Network Training: Fixed Point Analysis, Edge of Stability, and Route to Chaos [6.579523168465526]
  In descent dynamics of neural networks, the top eigenvalue of the Hessian of the loss (sharpness) displays a variety of robust phenomena throughout training. We demonstrate that a simple $2$-layer linear network (UV model) trained on a single training example exhibits all of the essential sharpness phenomenology observed in real-world scenarios.
  arXiv  Detail & Related papers  (2023-11-03T17:59:40Z)
• Anomaly Detection with Variance Stabilized Density Estimation [49.46356430493534]
  We present a variance-stabilized density estimation problem for maximizing the likelihood of the observed samples. To obtain a reliable anomaly detector, we introduce a spectral ensemble of autoregressive models for learning the variance-stabilized distribution. We have conducted an extensive benchmark with 52 datasets, demonstrating that our method leads to state-of-the-art results.
  arXiv  Detail & Related papers  (2023-06-01T11:52:58Z)
• Second-order regression models exhibit progressive sharpening to the edge of stability [30.92413051155244]
  We show that for quadratic objectives in two dimensions, a second-order regression model exhibits progressive sharpening towards a value that differs slightly from the edge of stability. In higher dimensions, the model generically shows similar behavior, even without the specific structure of a neural network.
  arXiv  Detail & Related papers  (2022-10-10T17:21:20Z)
• Heavy-tailed Streaming Statistical Estimation [58.70341336199497]
  We consider the task of heavy-tailed statistical estimation given streaming $p$ samples. We design a clipped gradient descent and provide an improved analysis under a more nuanced condition on the noise of gradients.
  arXiv  Detail & Related papers  (2021-08-25T21:30:27Z)
• Training Generative Adversarial Networks by Solving Ordinary Differential Equations [54.23691425062034]
  We study the continuous-time dynamics induced by GAN training. From this perspective, we hypothesise that instabilities in training GANs arise from the integration error. We experimentally verify that well-known ODE solvers (such as Runge-Kutta) can stabilise training.
  arXiv  Detail & Related papers  (2020-10-28T15:23:49Z)
• Fine-Grained Analysis of Stability and Generalization for Stochastic Gradient Descent [55.85456985750134]
  We introduce a new stability measure called on-average model stability, for which we develop novel bounds controlled by the risks of SGD iterates. This yields generalization bounds depending on the behavior of the best model, and leads to the first-ever-known fast bounds in the low-noise setting. To our best knowledge, this gives the firstever-known stability and generalization for SGD with even non-differentiable loss functions.
  arXiv  Detail & Related papers  (2020-06-15T06:30:19Z)
• Multiplicative noise and heavy tails in stochastic optimization [62.993432503309485]
  empirical optimization is central to modern machine learning, but its role in its success is still unclear. We show that it commonly arises in parameters of discrete multiplicative noise due to variance. A detailed analysis is conducted in which we describe on key factors, including recent step size, and data, all exhibit similar results on state-of-the-art neural network models.
  arXiv  Detail & Related papers  (2020-06-11T09:58:01Z)

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.