English
Related papers

Related papers: Strong inductive biases provably prevent harmless …

200 papers

This paper considers binary classification of high-dimensional features under a postulated model with a low-dimensional latent Gaussian mixture structure and non-vanishing noise. A generalized least squares estimator is used to estimate the…

Machine Learning · Statistics 2023-03-30 Xin Bing , Marten Wegkamp

Often the question arises whether $Y$ can be predicted based on $X$ using a certain model. Especially for highly flexible models such as neural networks one may ask whether a seemingly good prediction is actually better than fitting pure…

Methodology · Statistics 2024-04-30 Michał Ciszewski , Jakob Söhl , Ton Leenen , Bart van Trigt , Geurt Jongbloed

An evolving line of machine learning works observe empirical evidence that suggests interpolating estimators -- the ones that achieve zero training error -- may not necessarily be harmful. This paper pursues theoretical understanding for an…

Statistics Theory · Mathematics 2021-10-19 Yue Li , Yuting Wei

Fully convolutional neural networks can process input of arbitrary size by applying a combination of downsampling and pooling. However, we find that fully convolutional image classifiers are not agnostic to the input size but rather show…

Machine Learning · Computer Science 2021-10-13 Mats L. Richter , Wolf Byttner , Ulf Krumnack , Ludwdig Schallner , Justin Shenk

The practical success of deep learning has led to the discovery of several surprising phenomena. One of these phenomena, that has spurred intense theoretical research, is ``benign overfitting'': deep neural networks seem to generalize well…

Machine Learning · Computer Science 2026-02-25 Ichiro Hashimoto , Stanislav Volgushev , Piotr Zwiernik

How sensitive should machine learning models be to input changes? We tackle the question of model smoothness and show that it is a useful inductive bias which aids generalization, adversarial robustness, generative modeling and…

Machine Learning · Statistics 2021-07-08 Mihaela Rosca , Theophane Weber , Arthur Gretton , Shakir Mohamed

The phenomenon of benign overfitting, where a trained neural network perfectly fits noisy training data but still achieves near-optimal test performance, has been extensively studied in recent years for linear models and…

Machine Learning · Computer Science 2025-02-13 Roey Magen , Shuning Shang , Zhiwei Xu , Spencer Frei , Wei Hu , Gal Vardi

Spherical radial-basis-based kernel interpolation abounds in image sciences including geophysical image reconstruction, climate trends description and image rendering due to its excellent spatial localization property and perfect…

Machine Learning · Computer Science 2024-01-17 Xiaotong Liu , Jinxin Wang , Di Wang , Shao-Bo Lin

The kernel smoothing with large bandwidth values causes oversmoothing or underfitting in general. However, when irrelevant variables are included, the corresponding large bandwidth values are known to have an effect of shrinking them. This…

Statistics Theory · Mathematics 2026-03-05 Taku Moriyama

We study the generalization error of functions that interpolate prescribed data points and are selected by minimizing a weighted norm. Under natural and general conditions, we prove that both the interpolants and their generalization errors…

Numerical Analysis · Mathematics 2021-02-11 Weilin Li

The generalization ability of kernel interpolation in large dimensions (i.e., $n \asymp d^{\gamma}$ for some $\gamma>0$) might be one of the most interesting problems in the recent renaissance of kernel regression, since it may help us…

Machine Learning · Computer Science 2024-04-22 Haobo Zhang , Weihao Lu , Qian Lin

Most modern learning problems are over-parameterized, where the number of learnable parameters is much greater than the number of training data points. In this over-parameterized regime, the training loss typically has infinitely many…

Machine Learning · Computer Science 2025-06-23 Kanumuri Nithin Varma , Babak Hassibi

The remarkable practical success of deep learning has revealed some major surprises from a theoretical perspective. In particular, simple gradient methods easily find near-optimal solutions to non-convex optimization problems, and despite…

Statistics Theory · Mathematics 2021-03-17 Peter L. Bartlett , Andrea Montanari , Alexander Rakhlin

Pruning - that is, setting a significant subset of the parameters of a neural network to zero - is one of the most popular methods of model compression. Yet, several recent works have raised the issue that pruning may induce or exacerbate…

Computer Vision and Pattern Recognition · Computer Science 2023-04-26 Eugenia Iofinova , Alexandra Peste , Dan Alistarh

The objective function of a matrix factorization model usually aims to minimize the average of a regression error contributed by each element. However, given the existence of stochastic noises, the implicit deviations of sample data from…

Machine Learning · Computer Science 2016-10-31 Guang-He Lee , Shao-Wen Yang , Shou-De Lin

We analyse the interpolator with minimal $\ell_2$-norm $\hat{\beta}$ in a general high dimensional linear regression framework where $\mathbb Y=\mathbb X\beta^*+\xi$ where $\mathbb X$ is a random $n\times p$ matrix with independent…

Statistics Theory · Mathematics 2021-01-06 Geoffrey Chinot , Matthieu Lerasle

Both humans and large language models are able to learn language without explicit structural supervision. What inductive biases make this learning possible? We address this fundamental cognitive question by leveraging transformer language…

Computation and Language · Computer Science 2023-10-31 Isabel Papadimitriou , Dan Jurafsky

In all applications in digital communications, it is crucial for an estimator to be unbiased. Although so-called soft feedback is widely employed in many different fields of engineering, typically the biased estimate is used. In this paper,…

Information Theory · Computer Science 2018-02-21 Susanne Sparrer , Robert F. H. Fischer

Generalization performance of classifiers in deep learning has recently become a subject of intense study. Deep models, typically over-parametrized, tend to fit the training data exactly. Despite this "overfitting", they perform well on…

Machine Learning · Statistics 2018-06-18 Mikhail Belkin , Siyuan Ma , Soumik Mandal

Recent empirical and theoretical studies have established the generalization capabilities of large machine learning models that are trained to (approximately or exactly) fit noisy data. In this work, we prove a surprising result that even…

Machine Learning · Computer Science 2024-01-26 Yifan Hao , Tong Zhang