Detecting overfitting in Neural Networks during long-horizon grokking using Random Matrix Theory

• URL: http://arxiv.org/abs/2605.12394v2
• Date: Thu, 14 May 2026 04:36:50 GMT
• Title: Detecting overfitting in Neural Networks during long-horizon grokking using Random Matrix Theory
• Abstract summary: We present a novel Random Matrix Theory method that detects the onset of overfitting in deep learning models without access to train or test data.<n>During the onset of overfitting, which we call the "anti-grokking" phase in long-horizon grokking, Correlation Traps form and grow in number and scale.<n>Our findings show that anti-grokking is an additional grokking phase with high train accuracy and decreasing test accuracy, structurally distinct from pre-grokking through its Correlation Traps.
• Score: 1.6615337656760856
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Training Neural Networks (NNs) without overfitting is difficult; detecting that overfitting is difficult as well. We present a novel Random Matrix Theory method that detects the onset of overfitting in deep learning models without access to train or test data. For each model layer, we randomize each weight matrix element-wise, $\mathbf{W} \to \mathbf{W}^{\mathrm{rand}}$, fit the randomized empirical spectral distribution with a Marchenko-Pastur distribution, and identify large outliers that violate self-averaging. We call these outliers Correlation Traps. During the onset of overfitting, which we call the "anti-grokking" phase in long-horizon grokking, Correlation Traps form and grow in number and scale as test accuracy decreases while train accuracy remains high. Traps may be benign or may harm generalization; we provide an empirical approach to distinguish between them by passing random data through the trained model and evaluating the JS divergence of output logits. Our findings show that anti-grokking is an additional grokking phase with high train accuracy and decreasing test accuracy, structurally distinct from pre-grokking through its Correlation Traps. More broadly, we find that some foundation-scale LLMs exhibit the same Correlation Traps, indicating potentially harmful overfitting.

Related papers

• Late-Stage Generalization Collapse in Grokking: Detecting anti-grokking with Weightwatcher [1.6615337656760856]
  emphMemorization in neural networks lacks a precise operational definition and is often inferred from the grokking regime.<n>We identify a previously unreported third phase of grokking in this training regime: emphanti-grokking, a late-stage collapse of generalization.
  arXiv  Detail & Related papers  (2026-02-02T22:09:14Z)
• A Classical View on Benign Overfitting: The Role of Sample Size [14.36840959836957]
  We focus on almost benign overfitting, where models simultaneously achieve both arbitrarily small training and test errors.<n>This behavior is characteristic of neural networks, which often achieve low (but non-zero) training error while still generalizing well.
  arXiv  Detail & Related papers  (2025-05-16T18:37:51Z)
• Risk and cross validation in ridge regression with correlated samples [72.59731158970894]
  We provide training examples for the in- and out-of-sample risks of ridge regression when the data points have arbitrary correlations.<n>We demonstrate that in this setting, the generalized cross validation estimator (GCV) fails to correctly predict the out-of-sample risk.<n>We further extend our analysis to the case where the test point has nontrivial correlations with the training set, a setting often encountered in time series forecasting.
  arXiv  Detail & Related papers  (2024-08-08T17:27:29Z)
• Distributionally Robust Skeleton Learning of Discrete Bayesian Networks [9.46389554092506]
  We consider the problem of learning the exact skeleton of general discrete Bayesian networks from potentially corrupted data. We propose to optimize the most adverse risk over a family of distributions within bounded Wasserstein distance or KL divergence to the empirical distribution. We present efficient algorithms and show the proposed methods are closely related to the standard regularized regression approach.
  arXiv  Detail & Related papers  (2023-11-10T15:33:19Z)
• Robust computation of optimal transport by $\eta$-potential regularization [79.24513412588745]
  Optimal transport (OT) has become a widely used tool in the machine learning field to measure the discrepancy between probability distributions. We propose regularizing OT with the beta-potential term associated with the so-called $beta$-divergence. We experimentally demonstrate that the transport matrix computed with our algorithm helps estimate a probability distribution robustly even in the presence of outliers.
  arXiv  Detail & Related papers  (2022-12-26T18:37:28Z)
• Benign-Overfitting in Conditional Average Treatment Effect Prediction with Linear Regression [14.493176427999028]
  We study the benign overfitting theory in the prediction of the conditional average treatment effect (CATE) with linear regression models. We show that the T-learner fails to achieve the consistency except the random assignment, while the IPW-learner converges the risk to zero if the propensity score is known.
  arXiv  Detail & Related papers  (2022-02-10T18:51:52Z)
• Towards an Understanding of Benign Overfitting in Neural Networks [104.2956323934544]
  Modern machine learning models often employ a huge number of parameters and are typically optimized to have zero training loss. We examine how these benign overfitting phenomena occur in a two-layer neural network setting. We show that it is possible for the two-layer ReLU network interpolator to achieve a near minimax-optimal learning rate.
  arXiv  Detail & Related papers  (2021-06-06T19:08:53Z)
• Shaping Deep Feature Space towards Gaussian Mixture for Visual Classification [74.48695037007306]
  We propose a Gaussian mixture (GM) loss function for deep neural networks for visual classification. With a classification margin and a likelihood regularization, the GM loss facilitates both high classification performance and accurate modeling of the feature distribution. The proposed model can be implemented easily and efficiently without using extra trainable parameters.
  arXiv  Detail & Related papers  (2020-11-18T03:32:27Z)
• Good Classifiers are Abundant in the Interpolating Regime [64.72044662855612]
  We develop a methodology to compute precisely the full distribution of test errors among interpolating classifiers. We find that test errors tend to concentrate around a small typical value $varepsilon*$, which deviates substantially from the test error of worst-case interpolating model. Our results show that the usual style of analysis in statistical learning theory may not be fine-grained enough to capture the good generalization performance observed in practice.
  arXiv  Detail & Related papers  (2020-06-22T21:12:31Z)
• Uncertainty Estimation Using a Single Deep Deterministic Neural Network [66.26231423824089]
  We propose a method for training a deterministic deep model that can find and reject out of distribution data points at test time with a single forward pass. We scale training in these with a novel loss function and centroid updating scheme and match the accuracy of softmax models.
  arXiv  Detail & Related papers  (2020-03-04T12:27:36Z)

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.