arXiv Machine Learning

Floating-Point Networks with Automatic Differentiation Can Represent Almost All Floating-Point Functions and Their Gradients

arXiv:2605. 01702v2 Announce Type: replace Abstract: Theoretical studies show that for any differentiable function on a compact domain, there exists a neural network that approximates both the function values and gradients.

arXiv Machine Learning
Jul 16

Automatic Differentiation from Scratch: How PyTorch Computes Gradients in Physics-Informed Neural Networks

arXiv:2607. 13042v1 Announce Type: new Abstract: This paper traces, with explicit numerical values, how PyTorch's automatic differentiation (AD) engine computes gradients for Physics-Informed Neural Network (PINN) training -- a setting that requires two levels of differentiation: computing the physics derivative $\hat{y}'(t)=d\hat{y}/dt$ through the network, and computing parameter gradients $\nabla_\theta L$ of a loss that itself depends on $\hat{y}'(t)$.

By Abdeladhim Tahimi
arXiv Machine Learning
Jun 16

Functional Gradient Descent with Adaptive Representations

arXiv:2606. 16926v1 Announce Type: cross Abstract: Functional optimization problems are typically solved by optimizing the parameters of a fixed representation, such as a neural network, resulting in highly nonconvex losses that complicate both training and theoretical analysis.

By Daniel Csillag, Rodrigo Schuller, Pedro Dall'Antonia, Leonidas Guibas, Luiz Velho, Tiago Novello
Hugging Face Trending Papers
Aug 11

Derivative Computation in PINNs: Automatic Differentiation, Finite Differences and Beyond

We systematically investigate finite-difference (FD) derivative computation in Physics-Informed Neural Networks (PINNs) as an alternative to automatic differentiation (AD). On three benchmark PDEs we show that, with a properly calibrated step size, FD matches AD in accuracy on every problem while running faster across the full tested batch-size range and using substantially less GPU memory, and that a stochastic variant we propose outperforms AD on a stationary problem.

arXiv Machine Learning
Jul 14

Approximation of Analytic Functions by ReLU Neural Networks with Adjustable Depth and Width

arXiv:2607. 10589v1 Announce Type: cross Abstract: In contrast to most studies on neural network approximation theory that characterize results through a single parameter, such as the total number of network parameters, \cite{shen2020deep} pioneered the characterization of approximation rates as a joint function of the width parameter $N$ and the depth parameter $L$, thereby granting greater architectural flexibility.

By Yanming Lai, Defeng Sun, Yang Wang
arXiv Machine Learning
Jun 24

Layer-wise Geometric Approximation Rates for Deep Networks

arXiv:2604. 20219v2 Announce Type: replace Abstract: Depth is widely viewed as a central contributor to the success of deep neural networks, whereas standard neural network approximation theory typically provides guarantees only for the final output and leaves the role of intermediate layers largely unclear.

By Shijun Zhang, Zuowei Shen, Yuesheng Xu
arXiv Machine Learning
Jun 2

Multigrade Neural Network Approximation

arXiv:2601. 16884v3 Announce Type: replace Abstract: We study multigrade deep learning (MGDL) as a principled framework for structured error refinement in deep neural networks.

By Shijun Zhang, Zuowei Shen, Yuesheng Xu
arXiv Machine Learning
Jul 24

Compiling to recurrent neurons

arXiv:2511. 14953v2 Announce Type: replace-cross Abstract: Discrete structures are currently second-class in differentiable programming.

By Joey Velez-Ginorio, Nada Amin, Konrad Kording, Steve Zdancewic