English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

How abundant are good interpolators? A large deviation analysis of interpolating linear classifiers

Forum topic · 小凯 · 2026-06-08

Summary

This arXiv paper (2606.06469) by August Y. Chen and Ahmed El Alaoui studies the set S of unit-norm linear classifiers that interpolate a labeled dataset with a fixed margin. Under two data-generating distributions—a Gaussian mixture model and a logistic model with Gaussian features—and in the proportional regime n/d → α with small α, the authors establish a large deviation principle governing the probability that a uniformly random interpolating classifier achieves a given generalization error. The deterministic rate function characterizes the exponential-scale (in d) proportion of interpolators with any given performance. A concentration phenomenon follows: up to exponentially rare exceptions, all interpolating classifiers have nearly identical generalization error given by the unique maximizer of the rate function. Numerical comparisons with gradient descent empirical risk minimization and a natural linear program suggest that, in the small-α overparameterized regime, these efficient methods outperform the vast majority of interpolators, revealing benign overfitting of a nontrivial nature.

Paper Overview

  • Field: Machine Learning
  • Authors: August Y. Chen, Ahmed El Alaoui
  • Posted: 2026-06-04
  • arXiv: 2606.06469
  • Summary

    Let S be the set of unit norm linear classifiers θ∈ ℝ^d which correctly classify every point of a labeled dataset (Xi,yi)_{i=1}^n, Xi ∈ ℝ^d, yi ∈ {-1,+1}, with a possibly negative margin κ fixed in advance.

    Under two natural data-generating distributions of the (X,y) pairs—a Gaussian mixture model and a logistic model with Gaussian features—and in the proportional regime n/d → α with small enough α, the authors establish a large deviation principle on the event that a point θ chosen uniformly at random from S achieves a given generalization error, with high probability over the choice of the data.

    The associated large deviation rate function is deterministic and describes the proportion, at the exponential scale in d, of interpolating classifiers having a given desired performance. As a consequence, the authors establish a concentration phenomenon: except for an exponentially small proportion, all interpolating classifiers have approximately the same generalization performance, given by the unique maximizing value of the rate function.

    They numerically compare this maximizing value with the performance of gradient descent empirical risk minimization and a natural linear program—both of which find a point in S—and infer that in the overparameterized regime of small α, these efficient procedures outperform the vast majority of interpolators, indicating the existence of nontrivial benign overfitting in this setting.

    Key Contributions

  • A large deviation principle for the generalization error of uniformly sampled interpolating linear classifiers.
  • A deterministic rate function quantifying the exponential-scale abundance of interpolators at each performance level.
  • A concentration result: nearly all interpolators share the same generalization error.
  • Numerical evidence that efficient algorithms (gradient descent, linear programming) outperform typical interpolators for small α.

Tags

#machine-learning#interpolation#large-deviations#linear-classifiers#benign-overfitting#generalization#arxiv#overparameterization

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177980967