Accurate, Efficient and Scalable Training of Graph Neural Networks (2020-10-05T00:00:00.000000Z)

TL;DR

The proposed training outperforms the state-of-the-art in scalability, efficiency and accuracy simultaneously simultaneously and enables fast training of deeper GNNs, as demonstrated by orders of magnitude speedup compared to the Tensorflow implementation.

Authors

Hongkuan Zhou

3 papers

Ajitesh Srivastava

3 papers

R. Kannan

3 papers

V. Prasanna

3 papers

Hanqing Zeng

1 papers

TL;DR

Authors

References33 items

GraphACT: Accelerating GCN Training on CPU-FPGA Heterogeneous Platforms

KnightKing: a fast distributed graph random walk engine

GraphSAINT: Graph Sampling Based Inductive Learning Method

MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing

Multi-Label Image Recognition With Graph Convolutional Networks

Accurate, Efficient and Scalable Graph Embedding

Adaptive Sampling Towards Fast Graph Representation Learning

Graph Convolutional Neural Networks for Web-Scale Recommender Systems

FastGCN: Fast Learning with Graph Convolutional Networks via Importance Sampling

Graph Attention Networks

Stochastic Training of Graph Convolutional Networks with Variance Reduction

Stochastic Training of Graph Convolutional Networks

Accelerating PageRank using Partition-Centric Processing

Spatio-temporal Graph Convolutional Neural Network: A Deep Learning Framework for Traffic Forecasting

Inductive Representation Learning on Large Graphs

Reducing Pagerank Communication via Propagation Blocking

Exploring the Hidden Dimension in Graph Processing

On Large-Batch Training for Deep Learning: Generalization Gap and Sharp Minima

Semi-Supervised Classification with Graph Convolutional Networks

Benefits of Depth in Neural Networks

X-Stream: edge-centric graph processing using streaming partitions

Distributed Representations of Words and Phrases and their Compositionality

A Survey and Taxonomy of Graph Sampling

Matrix Chain Multiplication

Estimating and sampling graphs with multidimensional random walks

Kronecker Graphs: An Approach to Modeling Networks

Sampling from large graphs

Graphs over time: densification laws, shrinking diameters and possible explanations

Challenge!

An Efficient Method for Generating Discrete Random Variables with General Distributions

Accelerating GCN training on CPU-FPGA heterogeneous platforms, in: The 2020 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays, in: FPGA ’20

Dynamorio: Dynamic instrumentation tool platform

Snap datasets

Field of Study

Journal Information

Name

Volume