English
Related papers

Related papers: Primitive Recursion without Composition: Dynamical…

200 papers

Computation is classically studied in terms of automata, formal languages and algorithms; yet, the relation between neural dynamics and symbolic representations and operations is still unclear in traditional eliminative connectionism.…

Neural and Evolutionary Computing · Computer Science 2016-09-08 Giovanni Sirio Carmantini , Peter beim Graben , Mathieu Desroches , Serafim Rodrigues

Over-parameterized neural networks generalize well in practice without any explicit regularization. Although it has not been proven yet, empirical evidence suggests that implicit regularization plays a crucial role in deep learning and…

Machine Learning · Computer Science 2019-03-07 Masayoshi Kubo , Ryotaro Banno , Hidetaka Manabe , Masataka Minoji

Recurrent neural networks (RNNs) are a widely used tool for modeling sequential data, yet they are often treated as inscrutable black boxes. Given a trained recurrent network, we would like to reverse engineer it--to obtain a quantitative,…

Machine Learning · Computer Science 2019-12-06 Niru Maheswaranathan , Alex Williams , Matthew D. Golub , Surya Ganguli , David Sussillo

ReLU neural networks define piecewise linear functions of their inputs. However, initializing and training a neural network is very different from fitting a linear spline. In this paper, we expand empirically upon previous theoretical work…

Machine Learning · Statistics 2016-11-30 Kevin K. Chen , Anthony Gamst , Alden Walker

Equivariant neural networks have shown improved performance, expressiveness and sample complexity on symmetrical domains. But for some specific symmetries, representations, and choice of coordinates, the most common point-wise activations,…

Machine Learning · Computer Science 2024-01-18 Marco Pacini , Xiaowen Dong , Bruno Lepri , Gabriele Santin

To be effective in sequential data processing, Recurrent Neural Networks (RNNs) are required to keep track of past events by creating memories. While the relation between memories and the network's hidden state dynamics was established over…

Machine Learning · Computer Science 2019-09-17 Doron Haviv , Alexander Rivkind , Omri Barak

The massive computational costs of scaling modern deep learning architectures have driven the widespread use of parameter-efficient low-rank structures, such as LoRA and low-rank factorization. However, theoretical guarantees for their…

Machine Learning · Computer Science 2026-05-08 Ying Chen , Aoxi Li , Jihun Kim , Javad Lavaei

We study the realization map of deep ReLU networks, focusing on when a function determines its parameters up to scaling and permutation. To analyze hidden redundancies beyond these standard symmetries, we introduce a framework based on…

Machine Learning · Computer Science 2026-05-21 Moritz Grillo , Guido Montúfar

In this study, we establish that deep neural networks employing ReLU and ReLU$^2$ activation functions can effectively represent Lagrange finite element functions of any order on various simplicial meshes in arbitrary dimensions. We…

Numerical Analysis · Mathematics 2024-01-15 Juncai He , Jinchao Xu

Boolean circuits abstract away from physical details to focus on the logical structure and computational behaviour of digital components. Although such circuits have been studied for many decades, compositionality has been widely ignored or…

Logic in Computer Science · Computer Science 2026-03-24 Damian Arellanes

Studies of the dynamics of nonlinear recurrent neural networks often assume independent and identically distributed couplings, but large-scale connectomics data indicate that biological neural circuits exhibit markedly different…

Neurons and Cognition · Quantitative Biology 2025-10-28 David G. Clark , Owen Marschall , Alexander van Meegen , Ashok Litwin-Kumar

Conceptual hydrologic models remain the cornerstone of rainfall-runoff modeling, yet their calibration is often slow and numerically fragile. Most gradient-based parameter estimation methods rely on finite-difference approximations or…

Machine Learning · Computer Science 2026-02-09 Jasper A. Vrugt , Jonathan M. Frame , Ethan Bollman

Classical neural network approximation results take the form: for every function $f$ and every error tolerance $\epsilon > 0$, one constructs a neural network whose architecture and weights depend on $\epsilon$. This paper introduces a…

Neural and Evolutionary Computing · Computer Science 2025-11-20 Clemens Hutter , Valentin Abadie , Helmut Bölcskei

Neural ordinary differential equations (ODEs) have been attracting increasing attention in various research domains recently. There have been some works studying optimization issues and approximation capabilities of neural ODEs, but their…

Machine Learning · Computer Science 2022-03-04 Hanshu Yan , Jiawei Du , Vincent Y. F. Tan , Jiashi Feng

The sequential nature of autoregressive next-token prediction imposes a fundamental speed limit on large language models. While continuous flow models offer a path to parallel generation, they traditionally demand expensive iterative…

In a neural network with ReLU activations, the number of piecewise linear regions in the output can grow exponentially with depth. However, this is highly unlikely to happen when the initial parameters are sampled randomly, which therefore…

Machine Learning · Computer Science 2025-10-17 Max Milkert , David Hyde , Forrest Laine

The universal approximation property (UAP) holds a fundamental position in deep learning, as it provides a theoretical foundation for the expressive power of neural networks. It is widely recognized that a composition of linear and…

Systems and Control · Electrical Eng. & Systems 2025-04-01 Yifei Duan , Yongqiang Cai

A neural ordinary differential equation (neural ODE) is a machine learning model that is commonly described as a continuous-depth generalization of a residual network (ResNet) with a single residual block, or conversely, the ResNet can be…

Machine Learning · Computer Science 2025-10-14 Abdelrahman Sayed Sayed , Pierre-Jean Meyer , Mohamed Ghazel

On general regular simplicial partitions $\mathcal{T}$ of bounded polytopal domains $\Omega \subset \mathbb{R}^d$, $d\in\{2,3\}$, we construct \emph{exact neural network (NN) emulations} of all lowest order finite element spaces in the…

Numerical Analysis · Mathematics 2023-07-17 Marcello Longo , Joost A. A. Opschoor , Nico Disch , Christoph Schwab , Jakob Zech

Universal approximation theory offers a foundational framework to verify neural network expressiveness, enabling principled utilization in real-world applications. However, most existing theoretical constructions are established by…

Machine Learning · Computer Science 2026-01-27 ZeYu Li , ShiJun Zhang , TieYong Zeng , FengLei Fan