Representation Gap: Explaining the Unreasonable Effectiveness of Neural Networks from a Geometric Perspective

• URL: http://arxiv.org/abs/2605.21692v1
• Date: Wed, 20 May 2026 19:51:25 GMT
• Title: Representation Gap: Explaining the Unreasonable Effectiveness of Neural Networks from a Geometric Perspective
• Abstract summary: We introduce the Representation Gap, a metric closely related to the generalization error, but admitting better-behaved dynamics.<n>We derive a precise equivalent of the Representation Gap and show that it is governed by a single parameter, the textitintrinsic dimension of the task.<n>We show that this dynamic also extends to a broader range of tasks and training algorithms.
• Score: 1.8618013369281987
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Characterizing precisely the asymptotic generalization error of neural networks using parameters that can be estimated efficiently is a crucial problem in machine learning, which relies heavily on heuristics and practitioners' intuition to make key design choices. In order to mitigate this issue, we introduce the Representation Gap, a metric closely related to the generalization error, but admitting better-behaved asymptotic dynamics. Focusing on equivariant diffusion models and leveraging results from optimal quantization and point-process theory, we derive a precise asymptotic equivalent of the Representation Gap and show that it is governed by a single parameter, the \textit{intrinsic dimension} of the task, which is easy to interpret, efficient to estimate, and can be linked to the equivariances of common neural network architectures. We show that this asymptotic dynamic also extends to a broader range of tasks and training algorithms. Finally, we demonstrate empirically that our asymptotic law and intrinsic dimension estimation are accurate on a wide range of synthetic datasets, where these quantities are known, as well as on more realistic datasets, where we obtain results consistent with the related literature.

Related papers

• Nonparametric Identification and Inference for Counterfactual Distributions with Confounding [6.997978440999076]
  We propose nonparametric identification and semiparametric estimation of joint potential outcome in the presence of confounding.<n>By bridging classical semiparametric theory with modern representation learning, this work provides a robust statistical foundation for distributional and counterfactual inference in complex causal systems.
  arXiv  Detail & Related papers  (2026-02-17T05:00:13Z)
• Spatially-informed transformers: Injecting geostatistical covariance biases into self-attention for spatio-temporal forecasting [0.0]
  We propose a hybrid architecture that injects a geostatistic inductive bias directly into the decomposing self-attention mechanism via a learnable costatistics kernel.<n>We demonstrate the phenomenon of Deep Variography'', where the network successfully recovers the true spatial parameters of the underlying process end-to-end via backpropagation.
  arXiv  Detail & Related papers  (2025-12-19T15:32:24Z)
• Generalization of Neural Combinatorial Solvers Through the Lens of Adversarial Robustness [68.97830259849086]
  Most datasets only capture a simpler subproblem and likely suffer from spurious features. We study adversarial robustness - a local generalization property - to reveal hard, model-specific instances and spurious features. Unlike in other applications, where perturbation models are designed around subjective notions of imperceptibility, our perturbation models are efficient and sound. Surprisingly, with such perturbations, a sufficiently expressive neural solver does not suffer from the limitations of the accuracy-robustness trade-off common in supervised learning.
  arXiv  Detail & Related papers  (2021-10-21T07:28:11Z)
• Fractal Structure and Generalization Properties of Stochastic Optimization Algorithms [71.62575565990502]
  We prove that the generalization error of an optimization algorithm can be bounded on the complexity' of the fractal structure that underlies its generalization measure. We further specialize our results to specific problems (e.g., linear/logistic regression, one hidden/layered neural networks) and algorithms.
  arXiv  Detail & Related papers  (2021-06-09T08:05:36Z)
• Post-mortem on a deep learning contest: a Simpson's paradox and the complementary roles of scale metrics versus shape metrics [61.49826776409194]
  We analyze a corpus of models made publicly-available for a contest to predict the generalization accuracy of neural network (NN) models. We identify what amounts to a Simpson's paradox: where "scale" metrics perform well overall but perform poorly on sub partitions of the data. We present two novel shape metrics, one data-independent, and the other data-dependent, which can predict trends in the test accuracy of a series of NNs.
  arXiv  Detail & Related papers  (2021-06-01T19:19:49Z)
• Joint Network Topology Inference via Structured Fusion Regularization [70.30364652829164]
  Joint network topology inference represents a canonical problem of learning multiple graph Laplacian matrices from heterogeneous graph signals. We propose a general graph estimator based on a novel structured fusion regularization. We show that the proposed graph estimator enjoys both high computational efficiency and rigorous theoretical guarantee.
  arXiv  Detail & Related papers  (2021-03-05T04:42:32Z)
• Optimizing Variational Representations of Divergences and Accelerating their Statistical Estimation [6.34892104858556]
  Variational representations of divergences and distances between high-dimensional probability distributions offer significant theoretical insights. They have gained popularity in machine learning as a tractable and scalable approach for training probabilistic models. We develop a methodology for building new, tighter variational representations of divergences.
  arXiv  Detail & Related papers  (2020-06-15T21:32:21Z)
• 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)
• Asymptotic Analysis of an Ensemble of Randomly Projected Linear Discriminants [94.46276668068327]
  In [1], an ensemble of randomly projected linear discriminants is used to classify datasets. We develop a consistent estimator of the misclassification probability as an alternative to the computationally-costly cross-validation estimator. We also demonstrate the use of our estimator for tuning the projection dimension on both real and synthetic data.
  arXiv  Detail & Related papers  (2020-04-17T12:47:04Z)

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.