paper-with-me

Papers

Overparameterization and generalization error: weighted trigonometric interpolation

2020-06-15 · Yuege Xie, Hung-Hsu Chou, Holger Rauhut, Rachel Ward

Motivated by surprisingly good generalization properties of learned deep neural networks in overparameterized scenarios and by the related double descent phenomenon, this paper analyzes the relation between smoothness and low generalization error in an overparameterized linear learning problem. We study a random Fourier series model, where the task is to estimate the unknown Fourier coefficients from equidistant samples. We derive exact expressions for the generalization error of both plain and weighted least squares estimators. We show precisely how a bias towards smooth interpolants, in the form of weighted trigonometric interpolation, can lead to smaller generalization error in the overparameterized regime compared to the underparameterized regime. This provides insight into the power of overparameterization, which is common in modern machine learning.

📄 PDF Abstract BibTeX arXiv:2006.08495

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Weight Normalization Weight Normalization is a normalization method for training neural networks. It is inspired by batch normalization,…

Similar Papers 제목 키워드 기반

Double Descent and Other Interpolation Phenomena in GANs

2021-06-07 · Lorenzo Luzi, Yehuda Dar, Richard Baraniuk

We study overparameterization in generative adversarial networks (GANs) that can interpolate the training data. We show that overparameterization can improve generalization performance and accelerate the training process…

Double Descent and Overparameterization in Particle Physics Data

2025-09-01 · Matthias Vigl, Lukas Heinrich arxiv

Recently, the benefit of heavily overparameterized models has been observed in machine learning tasks: models with enough capacity to easily cross the \emph{interpolation threshold} improve in generalization error compar…

Harmless interpolation of noisy data in regression

2019-03-21 · Vidya Muthukumar, Kailas Vodrahalli, Vignesh Subramanian, Anant Sahai

A continuing mystery in understanding the empirical success of deep neural networks is their ability to achieve zero training error and generalize well, even when the training data is noisy and there are more parameters …

regression

Interpolation can hurt robust generalization even when there is no noise

2021-08-05 · NeurIPS 2021 12 · Konstantin Donhauser, Alexandru Ţifrea, Michael Aerni, Reinhard Heckel 외

Numerous recent works show that overparameterization implicitly reduces variance for min-norm interpolators and max-margin classifiers. These findings suggest that ridge regularization has vanishing benefits in high dime…

regression

Frozen Overparameterization: A Double Descent Perspective on Transfer Learning of Deep Neural Networks

2022-11-20 · Yehuda Dar, Lorenzo Luzi, Richard G. Baraniuk

We study the generalization behavior of transfer learning of deep neural networks (DNNs). We adopt the overparameterization perspective -- featuring interpolation of the training data (i.e., approximately zero train erro…

image-classificationImage ClassificationTransfer Learning