DATA-DRIVEN LEARNING OF GEOMETRIC SCATTERING MODULES FOR GNNS.

Alexander Tong, Frederick Wenkel, Kincaid Macdonald, Smita Krishnaswamy, Guy Wolf
{"title":"DATA-DRIVEN LEARNING OF GEOMETRIC SCATTERING MODULES FOR GNNS.","authors":"Alexander Tong, Frederick Wenkel, Kincaid Macdonald, Smita Krishnaswamy, Guy Wolf","doi":"10.1109/mlsp52302.2021.9596169","DOIUrl":null,"url":null,"abstract":"<p><p>We propose a new graph neural network (GNN) module, based on relaxations of recently proposed geometric scattering transforms, which consist of a cascade of graph wavelet filters. Our learnable geometric scattering (LEGS) module enables adaptive tuning of the wavelets to encourage band-pass features to emerge in learned representations. The incorporation of our LEGS-module in GNNs enables the learning of longer-range graph relations compared to many popular GNNs, which often rely on encoding graph structure via smoothness or similarity between neighbors. Further, its wavelet priors result in simplified architectures with significantly fewer learned parameters compared to competing GNNs. We demonstrate the predictive performance of LEGS-based networks on graph classification benchmarks, as well as the descriptive quality of their learned features in biochemical graph data exploration tasks.</p>","PeriodicalId":73290,"journal":{"name":"IEEE International Workshop on Machine Learning for Signal Processing : [proceedings]. IEEE International Workshop on Machine Learning for Signal Processing","volume":"2021 ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2021-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10026018/pdf/nihms-1829559.pdf","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE International Workshop on Machine Learning for Signal Processing : [proceedings]. IEEE International Workshop on Machine Learning for Signal Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/mlsp52302.2021.9596169","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2021/11/15 0:00:00","PubModel":"Epub","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

We propose a new graph neural network (GNN) module, based on relaxations of recently proposed geometric scattering transforms, which consist of a cascade of graph wavelet filters. Our learnable geometric scattering (LEGS) module enables adaptive tuning of the wavelets to encourage band-pass features to emerge in learned representations. The incorporation of our LEGS-module in GNNs enables the learning of longer-range graph relations compared to many popular GNNs, which often rely on encoding graph structure via smoothness or similarity between neighbors. Further, its wavelet priors result in simplified architectures with significantly fewer learned parameters compared to competing GNNs. We demonstrate the predictive performance of LEGS-based networks on graph classification benchmarks, as well as the descriptive quality of their learned features in biochemical graph data exploration tasks.

Abstract Image

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
以数据为驱动学习 gnns 的几何散射模块。
我们提出了一种新的图神经网络(GNN)模块,它基于最近提出的几何散射变换的松弛,由一串图小波滤波器组成。我们的可学习几何散射(LEGS)模块能对小波进行自适应调整,以鼓励在学习的表征中出现带通特征。与许多流行的 GNN 相比,在 GNN 中加入我们的 LEGS 模块能够学习更长距离的图关系,后者通常依赖于通过平滑度或邻域之间的相似性对图结构进行编码。此外,与同类 GNN 相比,其小波前验可简化架构,大大减少学习参数。我们展示了基于 LEGS 的网络在图分类基准上的预测性能,以及在生化图数据探索任务中学习到的特征的描述质量。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
DATA-DRIVEN LEARNING OF GEOMETRIC SCATTERING MODULES FOR GNNS. CONVOLUTIONAL RECURRENT NEURAL NETWORK BASED DIRECTION OF ARRIVAL ESTIMATION METHOD USING TWO MICROPHONES FOR HEARING STUDIES. LEARNING GENERAL TRANSFORMATIONS OF DATA FOR OUT-OF-SAMPLE EXTENSIONS. Statistical modelling and inference Probabilistic graphical models
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1