Multiresolution Reservoir Graph Neural Network

作     者：Pasa, Luca Navarin, Nicolo Sperduti, Alessandro 

作者机构：Univ Padua Human Inspired Technol Res Ctr HIT Dept Math Tullio Levi Civita I-35121 Padua Italy 

出 版 物：《IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS》 (IEEE Trans. Neural Networks Learn. Sys.)

年 卷 期：2022年第33卷第6期

页      面：2642-2653页

核心收录：

学科分类：0808[工学-电气工程] 08[工学] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：Department of Mathematics  University of Padua  through the SID/BIRD 2020 Project "Deep Graph Memory Networks" 

主　　题：Convolution Reservoirs Training Computational modeling Neural networks Graph neural networks Standards Deep learning graph neural network machine learning on graphs reservoir computing (RC) structured data 

摘      要：Graph neural networks are receiving increasing attention as state-of-the-art methods to process graph-structured data. However, similar to other neural networks, they tend to suffer from a high computational cost to perform training. Reservoir computing (RC) is an effective way to define neural networks that are very efficient to train, often obtaining comparable predictive performance with respect to the fully trained counterparts. Different proposals of reservoir graph neural networks have been proposed in the literature. However, their predictive performances are still slightly below the ones of fully trained graph neural networks on many benchmark datasets, arguably because of the oversmoothing problem that arises when iterating over the graph structure in the reservoir computation. In this work, we aim to reduce this gap defining a multiresolution reservoir graph neural network (MRGNN) inspired by graph spectral filtering. Instead of iterating on the nonlinearity in the reservoir and using a shallow readout function, we aim to generate an explicit k-hop unsupervised graph representation amenable for further, possibly nonlinear, processing. Experiments on several datasets from various application areas show that our approach is extremely fast and it achieves in most of the cases comparable or even higher results with respect to state-of-the-art approaches.