English
Related papers

Related papers: Benign Overfitting in Single-Head Attention

200 papers

Meta learning has demonstrated tremendous success in few-shot learning with limited supervised data. In those settings, the meta model is usually overparameterized. While the conventional statistical learning theory suggests that…

Machine Learning · Computer Science 2022-11-10 Lisha Chen , Songtao Lu , Tianyi Chen

Transfer learning is a key component of modern machine learning, enhancing the performance of target tasks by leveraging diverse data sources. Simultaneously, overparameterized models such as the minimum-$\ell_2$-norm interpolator (MNI) in…

Machine Learning · Statistics 2026-01-19 Yeichan Kim , Ilmun Kim , Seyoung Park

We study the overfitting behavior of fully connected deep Neural Networks (NNs) with binary weights fitted to perfectly classify a noisy training set. We consider interpolation using both the smallest NN (having the minimal number of…

Machine Learning · Computer Science 2024-10-28 Itamar Harel , William M. Hoza , Gal Vardi , Itay Evron , Nathan Srebro , Daniel Soudry

Benign overfitting refers to how over-parameterized neural networks can fit training data perfectly and generalize well to unseen data. While this has been widely investigated theoretically, existing works are limited to two-layer networks…

Machine Learning · Computer Science 2024-10-28 Shuning Shang , Xuran Meng , Yuan Cao , Difan Zou

Modern deep learning models are usually highly over-parameterized so that they can overfit the training data. Surprisingly, such overfitting neural networks can usually still achieve high prediction accuracy. To study this "benign…

Machine Learning · Computer Science 2023-10-04 Xuran Meng , Difan Zou , Yuan Cao

Modern machine learning models with a large number of parameters often generalize well despite perfectly interpolating noisy training data - a phenomenon known as benign overfitting. A foundational explanation for this in linear…

Machine Learning · Statistics 2025-11-18 Yuta Kondo

We obtain a sufficient condition for benign overfitting of linear regression problem. Our result does not rely on concentration argument but on small-ball assumption and thus can holds in heavy-tailed case. The basic idea is to establish a…

Statistics Theory · Mathematics 2021-01-05 Zong Shang

The widespread success of deep neural networks has revealed a surprise in classical machine learning: very complex models often generalize well while simultaneously overfitting training data. This phenomenon of benign overfitting has been…

Quantum Physics · Physics 2023-12-20 Evan Peters , Maria Schuld

Recent research in neural networks and machine learning suggests that using many more parameters than strictly required by the initial complexity of a regression problem can result in more accurate or faster-converging models -- contrary to…

Machine Learning · Computer Science 2023-05-18 Arthur Castello B. de Oliveira , Milad Siami , Eduardo D. Sontag

The recent success of neural network models has shone light on a rather surprising statistical phenomenon: statistical models that perfectly fit noisy data can generalize well to unseen test data. Understanding this phenomenon of…

Machine Learning · Statistics 2022-09-13 Niladri S. Chatterji , Philip M. Long , Peter L. Bartlett

Modern machine learning often operates in the regime where the number of parameters is much higher than the number of data points, with zero training loss and yet good generalization, thereby contradicting the classical bias-variance…

Machine Learning · Statistics 2021-02-08 Zhu Li , Weijie Su , Dino Sejdinovic

We investigate two causes for adversarial vulnerability in deep neural networks: bad data and (poorly) trained models. When trained with SGD, deep neural networks essentially achieve zero training error, even in the presence of label noise,…

Machine Learning · Computer Science 2020-07-09 Amartya Sanyal , Puneet K Dokania , Varun Kanade , Philip H. S. Torr

The success of over-parameterized neural networks trained to near-zero training error has caused great interest in the phenomenon of benign overfitting, where estimators are statistically consistent even though they interpolate noisy…

Machine Learning · Statistics 2024-11-07 Moritz Haas , David Holzmüller , Ulrike von Luxburg , Ingo Steinwart

Transformers serve as the foundational architecture for many successful large-scale models, demonstrating the ability to overfit the training data while maintaining strong generalization on unseen data, a phenomenon known as benign…

Machine Learning · Computer Science 2025-02-19 Yingying Zhang , Zhenyu Wu , Jian Li , Yong Liu

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

We study benign overfitting in two-layer ReLU networks trained using gradient descent and hinge loss on noisy data for binary classification. In particular, we consider linearly separable data for which a relatively small proportion of…

Machine Learning · Computer Science 2023-11-10 Erin George , Michael Murray , William Swartworth , Deanna Needell

``Benign overfitting'', the ability of certain algorithms to interpolate noisy training data and yet perform well out-of-sample, has been a topic of considerable recent interest. We show, using a fixed design setup, that an important class…

Machine Learning · Computer Science 2023-04-14 Daniel Beaglehole , Mikhail Belkin , Parthe Pandit

Overfitting in deep learning has been the focus of a number of recent works, yet its exact impact on the behavior of neural networks is not well understood. This study analyzes overfitting by examining how the distribution of logits alters…

Machine Learning · Computer Science 2019-10-02 Zeju Li , Konstantinos Kamnitsas , Ben Glocker

The widely observed 'benign overfitting phenomenon' in the neural network literature raises the challenge to the 'bias-variance trade-off' doctrine in the statistical learning theory. Since the generalization ability of the 'lazy trained'…

Machine Learning · Computer Science 2023-09-26 Yicheng Li , Haobo Zhang , Qian Lin

In many modern applications of deep learning the neural network has many more parameters than the data points used for its training. Motivated by those practices, a large body of recent theoretical research has been devoted to studying…

Statistics Theory · Mathematics 2022-12-07 A. Tsigler , P. L. Bartlett