Matching Convolutional Neural Networks without Priors about Data

Carlos Eduardo Rosar Kos Lassance,Jean-Charles Vialatte,Vincent Gripon

2018 IEEE Data Science Workshop (DSW)（2018）

引用 1|浏览0

暂无评分

摘要

We propose an extension of Convolutional Neural Networks (CNNs) to graph-structured data, including strided convolutions and data augmentation on graphs. Our method matches the accuracy of state-of-the-art CNNs when applied on images, without any prior about their 2D regular structure. On fMRI data, we obtain a significant gain in accuracy compared with existing graph-based alternatives.

查看译文

关键词

priors,strided convolutions,inferred graph translations,2D regular structure,fMRI data,graph-based alternatives,convolutional neural networks matching,data augmentation,graph-structured data

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要