paper-with-me

홈 › Papers

Chebyshev Feature Neural Network for Accurate Function Approximation

2024-09-27 · Zhongshu Xu, Yuan Chen, Dongbin Xiu

We present a new Deep Neural Network (DNN) architecture capable of approximating functions up to machine accuracy. Termed Chebyshev Feature Neural Network (CFNN), the new structure employs Chebyshev functions with learnable frequencies as the first hidden layer, followed by the standard fully connected hidden layers. The learnable frequencies of the Chebyshev layer are initialized with exponential distributions to cover a wide range of frequencies. Combined with a multi-stage training strategy, we demonstrate that this CFNN structure can achieve machine accuracy during training. A comprehensive set of numerical examples for dimensions up to $20$ are provided to demonstrate the effectiveness and scalability of the method.

📄 PDF Abstract BibTeX arXiv:2409.19135

Code (1)

AtticusXu/CFNN 공식 구현 tf

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Chebyshev Polynomial-Based Kolmogorov-Arnold Networks: An Efficient Architecture for Nonlinear Function Approximation

2024-05-12 · Sidharth SS, Keerthana AR, Gokul R, Anas KP

Accurate approximation of complex nonlinear functions is a fundamental challenge across many scientific and engineering domains. Traditional neural network architectures, such as Multi-Layer Perceptrons (MLPs), often str…

Kolmogorov-Arnold Networks

Beyond ReLU: Chebyshev-DQN for Enhanced Deep Q-Networks

2025-08-20 · Saman Yazdannik, Morteza Tayefi, Shamim Sanisales arxiv

The performance of Deep Q-Networks (DQN) is critically dependent on the ability of its underlying neural network to accurately approximate the action-value function. Standard function approximators, such as multi-layer p…

Reinforcement Learning

A Novel Online Pseudospectral Method for Approximation of Nonlinear Systems Dynamics

2025-05-12 · Arian Yousefian, Avimanyu Sahoo, Vignesh Narayanan

This paper presents a novel online system identification approach utilizing the Chebyshev pseudospectral (PS) method to approximate the dynamics of a continuous-time nonlinear system. Unlike conventional periodic samplin…

State Estimation

Convolutional Neural Networks on Graphs with Chebyshev Approximation, Revisited

2022-02-04 · Mingguo He, Zhewei Wei, Ji-Rong Wen

Designing spectral convolutional networks is a challenging problem in graph learning. ChebNet, one of the early attempts, approximates the spectral graph convolutions using Chebyshev polynomials. GCN simplifies ChebNet b…

GPRGraph LearningNode Classification

Iterative Polynomial Approximation Algorithms for Inverse Graph Filters

2025-04-19 · Cheng Cheng, Qiyu Sun, Cong Zheng

Chebyshev interpolation polynomials exhibit the exponential approximation property to analytic functions on a cube. Based on the Chebyshev interpolation polynomial approximation, we propose iterative polynomial approxima…