12 datasets found

The ogbn-papers100M dataset is a benchmark for graph convolutional networks.

Graph Neural Networks (GNNs) have become a popular approach for various applications, ranging from social network analysis to modeling chemical properties of molecules.

The relational data model was designed to facilitate large-scale data management and analytics. We consider the problem of how to differentiate computations expressed relationally.

Simplifying graph convolutional networks.

Graph Convolutional Networks (GCNs) with multi-hop aggregation is more expressive than one-hop GCN but suffers from higher model complexity. Finding the shortest aggregation...

This paper aims at studying the sample complexity of graph convolutional neural networks (GCNs), by providing tight upper bounds of Rademacher complexity for GCN models with a...

Graph convolutional networks are a popular class of deep neural network algorithms which have shown success in a number of relational learning tasks. Despite their success,...

Distance-geometric graph representation for 3D graphs, utilizing continuous-filter convolutional layers with filter-generating networks.

Experimental results of DG-GAT for the ESOL and FreeSolv datasets show major improvement over those of the standard graph convolution network based on 2D molecular graphs.

The dataset used in the paper is a generic framework for convolution on arbitrary structures, which includes grid convolutions and graph convolutions.

Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecasting

A dataset of small molecules for benchmarking molecule generation methods. The dataset consists of fingerprints of the molecules, and the goal is to predict the original...