Grokking as a Variance-Limited Phase Transition: Spectral Gating and the Epsilon-Stability Threshold

• URL: http://arxiv.org/abs/2603.15492v1
• Date: Mon, 16 Mar 2026 16:18:18 GMT
• Title: Grokking as a Variance-Limited Phase Transition: Spectral Gating and the Epsilon-Stability Threshold
• Abstract summary: Generalization requires the textitanisotropic rectification unique to adaptives, which directs noise into the tangent space of the solution.<n>This work analyzes AdamW dynamics on modular arithmetic tasks, revealing a Spectral Gating'' mechanism that regulates the transition from generalization to memorization.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Standard optimization theories struggle to explain grokking, where generalization occurs long after training convergence. While geometric studies attribute this to slow drift, they often overlook the interaction between the optimizer's noise structure and landscape curvature. This work analyzes AdamW dynamics on modular arithmetic tasks, revealing a ``Spectral Gating'' mechanism that regulates the transition from memorization to generalization. We find that AdamW operates as a variance-gated stochastic system. Grokking is constrained by a stability condition: the generalizing solution resides in a sharp basin ($λ_{max}^H$) initially inaccessible under low-variance regimes. The ``delayed'' phase represents the accumulation of gradient variance required to lift the effective stability ceiling, permitting entry into this sharp manifold. Our ablation studies identify three complexity regimes: (1) \textbf{Capacity Collapse} ($P < 23$), where rank-deficiency prevents structural learning; (2) \textbf{The Variance-Limited Regime} ($P \approx 41$), where generalization waits for the spectral gate to open; and (3) \textbf{Stability Override} ($P > 67$), where memorization becomes dimensionally unstable. Furthermore, we challenge the "Flat Minima" hypothesis for algorithmic tasks, showing that isotropic noise injection fails to induce grokking. Generalization requires the \textit{anisotropic rectification} unique to adaptive optimizers, which directs noise into the tangent space of the solution manifold.

Related papers

• Phasor Attention: Mean Root Square Normalization for Phase Manifold Preservation [3.0459410574367314]
  We introduce Mean Root Square Normalization (MRSNorm)<n>MRSNorm mathematically inverts the traditional scaling paradigm.<n>We analytically demonstrate that this constraint yields a built-in, trigonometric gradient clipper governed by the Pythagorean identity.
  arXiv  Detail & Related papers  (2026-07-20T11:10:51Z)
• Sample Complexity of Policy Gradient for Log-Growth Control [23.86784769848408]
  We study the sample complexity of policy gradient for log-growth control.<n>We prove that projected mini-batch policy gradient attains total sample complexity.
  arXiv  Detail & Related papers  (2026-05-26T07:24:14Z)
• Exact Finite-Sample Variance Decomposition of Subagging: A Spectral Filtering Perspective [17.68593329351594]
  We use the Hoeffding-ANOVA decomposition to derive the first exact, finite-sample variance decomposition for subagging.<n>We show that subagging preserves low-order structural signals while atten $c$-th order interaction variance by a geometric factor approaching $c$.<n>This decoupling reveals why default baselines often under-regularize high-capacity interpolators, which instead require smaller $$ to exponentially suppress spurious high-order noise.
  arXiv  Detail & Related papers  (2026-04-12T05:47:41Z)
• A Systematic Empirical Study of Grokking: Depth, Architecture, Activation, and Regularization [0.0]
  Grokking the delayed transition from memorization to generalization in neural networks remains poorly understood.<n>We present a controlled study that systematically disentangles factors on modular addition (mod 97)<n>Our central finding is that grokking dynamics are not primarily determined by architecture, but by interactions between stability and regularization.
  arXiv  Detail & Related papers  (2026-03-26T04:16:01Z)
• OptEMA: Adaptive Exponential Moving Average for Stochastic Optimization with Zero-Noise Optimality [23.28384210732827]
  We introduce OptEMA and analyze two novel variants: OptEMA-M and OptEMA-V.<n>OptEMA is closed-loop and Lipschitz-free in the sense that its effective stepsizes are trajectory-dependent and do not require the Lipschitz constant for parameterization.<n>Both variants achieve a noise-adaptive convergence rate of $widetildemathcalO(T-1/2+1/2 T-1/4)$ for the average gradient norm, where $$ is the noise level.
  arXiv  Detail & Related papers  (2026-03-10T17:19:54Z)
• Stability and Generalization of Push-Sum Based Decentralized Optimization over Directed Graphs [55.77845440440496]
  Push-based decentralized communication enables optimization over communication networks, where information exchange may be asymmetric.<n>We develop a unified uniform-stability framework for the Gradient Push (SGP) algorithm.<n>A key technical ingredient is an imbalance-aware generalization bound through two quantities.
  arXiv  Detail & Related papers  (2026-02-24T05:32:03Z)
• Why is Normalization Preferred? A Worst-Case Complexity Theory for Stochastically Preconditioned SGD under Heavy-Tailed Noise [17.899443444882888]
  We develop a worst-case complexity theory for inequalityally preconditioned gradient descent (SPSGD)<n>We demonstrate that normalization guarantees convergence to a first-order stationary point at rate $mathcalO(T-fracp-13p-2)$ when problem parameters are known, and $mathcalO(T-fracp-12p)$ when problem parameters are unknown.<n>In contrast, we prove that clipping may fail to converge in the worst case due to the statistical dependence between the preconditioner and the gradient estimates.
  arXiv  Detail & Related papers  (2026-02-13T19:29:17Z)
• Analysis of Hessian Scaling for Local and Global Costs in Variational Quantum Algorithm [0.42970700836450487]
  We quantify the entrywise resolvability of the Hessian for Variational Quantum Algorithms.<n>We show two distinct scaling regimes that govern the sample complexity required to resolve Hessian entries against shot noise.
  arXiv  Detail & Related papers  (2026-01-31T15:49:23Z)
• Latent Object Permanence: Topological Phase Transitions, Free-Energy Principles, and Renormalization Group Flows in Deep Transformer Manifolds [0.5729426778193398]
  We study the emergence of multi-step reasoning in deep Transformer language models through a geometric and statistical-physics lens.<n>We formalize the forward pass as a discrete coarse-graining map and relate the appearance of stable "concept basins" to fixed points of this renormalization-like dynamics.<n>The resulting low-entropy regime is characterized by a spectral tail collapse and by the formation of transient, reusable object-like structures in representation space.
  arXiv  Detail & Related papers  (2026-01-16T23:11:02Z)
• Estimation of Toeplitz Covariance Matrices using Overparameterized Gradient Descent [1.7188280334580195]
  We revisit Toeplitz covariance estimation through the lens of simple descent (GD)<n>We show that when $K = P$, GD may converge to suboptimal solutions.<n>We propose an accelerated GD variant with separate learning rates for amplitudes and frequencies.
  arXiv  Detail & Related papers  (2025-11-03T14:07:53Z)
• Revisiting Zeroth-Order Optimization: Minimum-Variance Two-Point Estimators and Directionally Aligned Perturbations [57.179679246370114]
  We identify the distribution of random perturbations that minimizes the estimator's variance as the perturbation stepsize tends to zero.<n>Our findings reveal that such desired perturbations can align directionally with the true gradient, instead of maintaining a fixed length.
  arXiv  Detail & Related papers  (2025-10-22T19:06:39Z)
• 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)
• Learning Globally Smooth Functions on Manifolds [94.22412028413102]
  Learning smooth functions is generally challenging, except in simple cases such as learning linear or kernel models. This work proposes to overcome these obstacles by combining techniques from semi-infinite constrained learning and manifold regularization. We prove that, under mild conditions, this method estimates the Lipschitz constant of the solution, learning a globally smooth solution as a byproduct.
  arXiv  Detail & Related papers  (2022-10-01T15:45:35Z)
• Improved Convergence Rate of Stochastic Gradient Langevin Dynamics with Variance Reduction and its Application to Optimization [50.83356836818667]
  gradient Langevin Dynamics is one of the most fundamental algorithms to solve non-eps optimization problems. In this paper, we show two variants of this kind, namely the Variance Reduced Langevin Dynamics and the Recursive Gradient Langevin Dynamics.
  arXiv  Detail & Related papers  (2022-03-30T11:39:00Z)
• Benign Overfitting of Constant-Stepsize SGD for Linear Regression [122.70478935214128]
  inductive biases are central in preventing overfitting empirically. This work considers this issue in arguably the most basic setting: constant-stepsize SGD for linear regression. We reflect on a number of notable differences between the algorithmic regularization afforded by (unregularized) SGD in comparison to ordinary least squares.
  arXiv  Detail & Related papers  (2021-03-23T17:15:53Z)
• Balancing Rates and Variance via Adaptive Batch-Size for Stochastic Optimization Problems [120.21685755278509]
  In this work, we seek to balance the fact that attenuating step-size is required for exact convergence with the fact that constant step-size learns faster in time up to an error. Rather than fixing the minibatch the step-size at the outset, we propose to allow parameters to evolve adaptively.
  arXiv  Detail & Related papers  (2020-07-02T16:02:02Z)
• 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.