English
Related papers

Related papers: Infinite-Width Limit of a Single Attention Layer: …

200 papers

In modern deep learning, there is a recent and growing literature on the interplay between large-width asymptotic properties of deep Gaussian neural networks (NNs), i.e. deep NNs with Gaussian-distributed weights, and Gaussian stochastic…

Machine Learning · Computer Science 2022-06-27 Stefano Favaro , Sandra Fortini , Stefano Peluchetti

There has been a recent surge of interest in modeling neural networks (NNs) as Gaussian processes. In the limit of a NN of infinite width the NN becomes equivalent to a Gaussian process. Here we demonstrate that for an ensemble of large,…

Machine Learning · Computer Science 2019-08-28 Joseph M. Antognini

Transductive tasks on graphs differ fundamentally from typical supervised machine learning tasks, as the independent and identically distributed (i.i.d.) assumption does not hold among samples. Instead, all train/test/validation samples are…

Machine Learning · Computer Science 2024-11-21 Hamed Shirzad , Honghao Lin , Ameya Velingker , Balaji Venkatachalam , David Woodruff , Danica Sutherland

Gaussian processes are ubiquitous in nature and engineering. A case in point is a class of neural networks in the infinite-width limit, whose priors correspond to Gaussian processes. Here we perturbatively extend this correspondence to…

Machine Learning · Statistics 2020-08-28 Sho Yaida

This paper investigates the approximation power of three types of random neural networks: (a) infinite width networks, with weights following an arbitrary distribution; (b) finite width networks obtained by subsampling the preceding…

Machine Learning · Computer Science 2019-10-21 Bolton Bailey , Ziwei Ji , Matus Telgarsky , Ruicheng Xian

While Bayesian neural networks (BNNs) hold the promise of being flexible, well-calibrated statistical models, inference often requires approximations whose consequences are poorly understood. We study the quality of common variational…

Machine Learning · Statistics 2020-10-26 Andrew Y. K. Foong , David R. Burt , Yingzhen Li , Richard E. Turner

In this paper, we provide the first precise distributional characterization of gradient descent iterates for general multi-layer neural networks under the canonical single-index regression model, in the `finite-width proportional regime'…

Machine Learning · Computer Science 2025-05-09 Qiyang Han , Masaaki Imaizumi

We study the approximation gap between the dynamics of a polynomial-width neural network and its infinite-width counterpart, both trained using projected gradient descent in the mean-field scaling regime. We demonstrate how to tightly bound…

Machine Learning · Statistics 2025-09-25 Margalit Glasgow , Denny Wu , Joan Bruna

We consider the infinite-width limit of a fully connected deep neural network with general weights, and we prove quantitative general bounds on the $2$-Wasserstein distance between the network and its infinite-width Gaussian limit, under…

Probability · Mathematics 2026-05-05 Filippo Giovagnini , Sotirios Kotitsas , Marco Romito

Using entropic inequalities from information theory, we provide new bounds on the total variation and 2-Wasserstein distances between a conditionally Gaussian law and a Gaussian law with invertible covariance matrix. We apply our results to…

Probability · Mathematics 2025-06-04 Lucia Celli , Giovanni Peccati

In this paper, we study approximation properties of single hidden layer neural networks with weights varying on finitely many directions and thresholds from an open interval. We obtain a necessary and at the same time sufficient measure…

Machine Learning · Computer Science 2023-04-05 Vugar Ismailov , Ekrem Savas

We propose a neural-network construction of Euclidean scalar quantum field theories from transformer attention heads, defining $n$-point correlators by averaging over random network parameters in the NN-QFT framework. For a single attention…

Machine Learning · Computer Science 2026-02-12 Dmitry S. Ageev , Yulia A. Ageeva

Infinitely wide or deep neural networks (NNs) with independent and identically distributed (i.i.d.) parameters have been shown to be equivalent to Gaussian processes. Because of the favorable properties of Gaussian processes, this…

Machine Learning · Computer Science 2026-03-24 Steven Adams , Andrea Patanè , Morteza Lahijanian , Luca Laurenti

We study signal propagation in linear recurrent models at finite width. While existing signal propagation theory relies predominantly on the infinite-width limit, it remains unclear for how long that approximation remains accurate when…

Machine Learning · Computer Science 2026-05-07 Mariia Seleznova

Neural networks with wide layers have attracted significant attention due to their equivalence to Gaussian processes, enabling perfect fitting of training data while maintaining generalization performance, known as benign overfitting.…

Machine Learning · Computer Science 2023-10-18 Tianxiang Gao , Xiaokai Huo , Hailiang Liu , Hongyang Gao

While linear attention architectures offer efficient inference, compressing unbounded history into a fixed-size memory inherently limits expressivity and causes information loss. To address this limitation, we introduce Random Access Memory…

Machine Learning · Computer Science 2026-02-13 Kaicheng Xiao , Haotian Li , Liran Dong , Guoliang Xing

We give a proof that, under relatively mild conditions, fully-connected feed-forward deep random neural networks converge to a Gaussian mixture distribution as only the width of the last hidden layer goes to infinity. We conducted…

Machine Learning · Statistics 2022-04-27 Yasuhiko Asao , Ryotaro Sakamoto , Shiro Takagi

Trained attention layers exhibit striking and reproducible spectral structure of the weights, including low-rank collapse, bulk deformation, and isolated spectral outliers, yet the origin of these phenomena and their implications for…

The interplay between infinite-width neural networks (NNs) and classes of Gaussian processes (GPs) is well known since the seminal work of Neal (1996). While numerous theoretical refinements have been proposed in the recent years, the…

Machine Learning · Statistics 2024-11-18 Daniele Bracale , Stefano Favaro , Sandra Fortini , Stefano Peluchetti

The Universal Approximation Theorem posits that neural networks can theoretically possess unlimited approximation capacity with a suitable activation function and a freely chosen or trained set of parameters. However, a more practical…

Machine Learning · Computer Science 2024-09-26 Li Liu , Tengchao Yu , Heng Yong