English
Related papers

Related papers: Understanding Generalization in Transformers: Erro…

200 papers

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

High complexity models are notorious in machine learning for overfitting, a phenomenon in which models well represent data but fail to generalize an underlying data generating process. A typical procedure for circumventing overfitting…

Machine Learning · Statistics 2025-03-11 James Schmidt

Generalization is one of the fundamental issues in machine learning. However, traditional techniques like uniform convergence may be unable to explain generalization under overparameterization. As alternative approaches, techniques based on…

Machine Learning · Computer Science 2022-03-22 Jiaye Teng , Jianhao Ma , Yang Yuan

Adversarial training is a widely used method to improve the robustness of deep neural networks (DNNs) over adversarial perturbations. However, it is empirically observed that adversarial training on over-parameterized networks often suffers…

Machine Learning · Statistics 2024-01-25 Zhongjie Shi , Fanghui Liu , Yuan Cao , Johan A. K. Suykens

A major challenge for transformers is generalizing to sequences longer than those observed during training. While previous works have empirically shown that transformers can either succeed or fail at length generalization depending on the…

Machine Learning · Computer Science 2025-05-01 Xinting Huang , Andy Yang , Satwik Bhattamishra , Yash Sarrof , Andreas Krebs , Hattie Zhou , Preetum Nakkiran , Michael Hahn

We provide an information-theoretic analysis of the generalization ability of Gibbs-based transfer learning algorithms by focusing on two popular transfer learning approaches, $\alpha$-weighted-ERM and two-stage-ERM. Our key result is an…

Machine Learning · Computer Science 2021-11-03 Yuheng Bu , Gholamali Aminian , Laura Toni , Miguel Rodrigues , Gregory Wornell

The remarkable capability of over-parameterised neural networks to generalise effectively has been explained by invoking a ``simplicity bias'': neural networks prevent overfitting by initially learning simple classifiers before progressing…

Computation and Language · Computer Science 2025-10-02 Riccardo Rende , Federica Gerace , Alessandro Laio , Sebastian Goldt

This paper focuses on over-parameterized deep neural networks (DNNs) with ReLU activation functions and proves that when the data distribution is well-separated, DNNs can achieve Bayes-optimal test error for classification while obtaining…

Machine Learning · Computer Science 2023-06-01 Zhenyu Zhu , Fanghui Liu , Grigorios G Chrysos , Francesco Locatello , Volkan Cevher

Although numerous methods to reduce the overfitting of convolutional neural networks (CNNs) exist, it is still not clear how to confidently measure the degree of overfitting. A metric reflecting the overfitting level might be, however,…

Machine Learning · Computer Science 2022-09-28 Svetlana Pavlitskaya , Joël Oswald , J. Marius Zöllner

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

This paper explores the connection between learning trajectories of Deep Neural Networks (DNNs) and their generalization capabilities when optimized using (stochastic) gradient descent algorithms. Instead of concentrating solely on the…

Machine Learning · Computer Science 2023-11-01 Jingwen Fu , Zhizheng Zhang , Dacheng Yin , Yan Lu , Nanning Zheng

Length generalization is a key property of a learning algorithm that enables it to make correct predictions on inputs of any length, given finite training data. To provide such a guarantee, one needs to be able to compute a length…

Machine Learning · Computer Science 2026-03-04 Andy Yang , Pascal Bergsträßer , Georg Zetzsche , David Chiang , Anthony W. Lin

Will further scaling up of machine learning models continue to bring success? A significant challenge in answering this question lies in understanding generalization gap, which is the impact of overfitting. Understanding generalization gap…

Machine Learning · Statistics 2026-05-18 Atsushi Suzuki , Jing Wang

Calibration is a fundamental property of a good predictive model: it requires that the model predicts correctly in proportion to its confidence. Modern neural networks, however, provide no strong guarantees on their calibration -- and can…

Machine Learning · Computer Science 2022-10-07 A. Michael Carrell , Neil Mallinar , James Lucas , Preetum Nakkiran

In supervised learning, it is known that overparameterized neural networks with one hidden layer provably and efficiently learn and generalize, when trained using stochastic gradient descent with a sufficiently small learning rate and…

Machine Learning · Computer Science 2022-03-24 Kulin Shah , Amit Deshpande , Navin Goyal

Margin enlargement over training data has been an important strategy since perceptrons in machine learning for the purpose of boosting the robustness of classifiers toward a good generalization ability. Yet Breiman (1999) showed a dilemma…

Machine Learning · Computer Science 2021-01-05 Weizhi Zhu , Yifei Huang , Yuan Yao

Generalization remains a central yet unresolved challenge in deep learning, particularly the ability to predict a model's performance beyond its training distribution using quantities available prior to test-time evaluation. Building on the…

Despite their widespread use, training deep Transformers can be unstable. Layer normalization, a standard component, improves training stability, but its placement has often been ad-hoc. In this paper, we conduct a principled study on the…

Machine Learning · Computer Science 2025-10-14 Kelvin Kan , Xingjian Li , Benjamin J. Zhang , Tuhin Sahai , Stanley Osher , Krishna Kumar , Markos A. Katsoulakis

In this work, we analyze various scaling limits of the training dynamics of transformer models in the feature learning regime. We identify the set of parameterizations that admit well-defined infinite width and depth limits, allowing the…

Machine Learning · Statistics 2024-10-07 Blake Bordelon , Hamza Tahir Chaudhry , Cengiz Pehlevan

Generalization of deep neural networks remains one of the main open problems in machine learning. Previous theoretical works focused on deriving tight bounds of model complexity, while empirical works revealed that neural networks exhibit…

Machine Learning · Computer Science 2022-01-31 James Wang , Cheng-Lin Yang