BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation

06/21/2021
by   Mingguo He, et al.
0

Many representative graph neural networks, e.g., GPR-GNN and ChebyNet, approximate graph convolutions with graph spectral filters. However, existing work either applies predefined filter weights or learns them without necessary constraints, which may lead to oversimplified or ill-posed filters. To overcome these issues, we propose BernNet, a novel graph neural network with theoretical support that provides a simple but effective scheme for designing and learning arbitrary graph spectral filters. In particular, for any filter over the normalized Laplacian spectrum of a graph, our BernNet estimates it by an order-K Bernstein polynomial approximation and designs its spectral property by setting the coefficients of the Bernstein basis. Moreover, we can learn the coefficients (and the corresponding filter weights) based on observed graphs and their associated signals and thus achieve the BernNet specialized for the data. Our experiments demonstrate that BernNet can learn arbitrary spectral filters, including complicated band-rejection and comb filters, and it achieves superior performance in real-world graph modeling tasks.

READ FULL TEXT
research
09/29/2020

Framework for Designing Filters of Spectral Graph Convolutional Neural Networks in the Context of Regularization Theory

Graph convolutional neural networks (GCNNs) have been widely used in gra...
research
02/04/2022

Convolutional Neural Networks on Graphs with Chebyshev Approximation, Revisited

Designing spectral convolutional networks is a challenging problem in gr...
research
07/18/2017

Spectral Filter Tracking

Visual object tracking is a challenging computer vision task with numero...
research
10/07/2021

Direct design of biquad filter cascades with deep learning by sampling random polynomials

Designing infinite impulse response filters to match an arbitrary magnit...
research
03/01/2022

When A Conventional Filter Meets Deep Learning: Basis Composition Learning on Image Filters

Image filters are fast, lightweight and effective, which make these conv...
research
07/16/2023

Automated Polynomial Filter Learning for Graph Neural Networks

Polynomial graph filters have been widely used as guiding principles in ...
research
02/24/2023

Graph Neural Networks with Learnable and Optimal Polynomial Bases

Polynomial filters, a kind of Graph Neural Networks, typically use a pre...

Please sign up or login with your details

Forgot password? Click here to reset