paper-with-me

홈 › Papers

Attention Mechanism, Max-Affine Partition, and Universal Approximation

2025-04-28 · Hude Liu, Jerry Yao-Chieh Hu, Zhao Song, Han Liu

We establish the universal approximation capability of single-layer, single-head self- and cross-attention mechanisms with minimal attached structures. Our key insight is to interpret single-head attention as an input domain-partition mechanism that assigns distinct values to subregions. This allows us to engineer the attention weights such that this assignment imitates the target function. Building on this, we prove that a single self-attention layer, preceded by sum-of-linear transformations, is capable of approximating any continuous function on a compact domain under the $L_\infty$-norm. Furthermore, we extend this construction to approximate any Lebesgue integrable function under $L_p$-norm for $1\leq p <\infty$. Lastly, we also extend our techniques and show that, for the first time, single-head cross-attention achieves the same universal approximation guarantees.

📄 PDF Abstract BibTeX arXiv:2504.19901

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Learning Theory of Transformers: Local-to-Global Approximation via Softmax Partition of Unity

2026-05-09 · Zhongjie Shi, Wenjing Liao arxiv

This paper investigates the learning theory of Transformer networks for regression tasks on the compact Euclidean domain $[0,1]^d$ and $d$-dimensional compact Riemannian manifolds. We propose a novel constructive approxi…

Necessary and sufficient conditions for universality of Kolmogorov-Arnold networks

2026-04-26 · Vugar Ismailov arxiv

We analyze the universal approximation property of Kolmogorov-Arnold Networks (KANs) in terms of their edge functions. If these functions are all affine, then universality clearly fails. How many non-affine functions are…

The universal approximation power of finite-width deep ReLU networks

2018-06-05 · ICLR 2019 5 · Dmytro Perekrestenko, Philipp Grohs, Dennis Elbrächter, Helmut Bölcskei

We show that finite-width deep ReLU neural networks yield rate-distortion optimal approximation (B\"olcskei et al., 2018) of polynomials, windowed sinusoidal functions, one-dimensional oscillatory textures, and the Weier…

Training-Time Batch Normalization Reshapes Local Partition Geometry in Piecewise-Affine Networks

2026-05-06 · Xuan Qi, Yi Wei, Fanqi Yu, Furao Shen 외 arxiv

Batch normalization (BN) is central to modern deep networks, but its effect on the realized function during training remains less understood than its optimization benefits. We study training-time BN in continuous piecewi…

On the Omnipresence of Spurious Local Minima in Certain Neural Network Training Problems

2022-02-23 · Constantin Christof, Julia Kowalczyk

We study the loss landscape of training problems for deep artificial neural networks with a one-dimensional real output whose activation functions contain an affine segment and whose hidden layers have width at least two…