TY - GEN

T1 - Graph structure of neural networks

AU - You, Jiaxuan

AU - Leskovec, Jure

AU - He, Kaiming

AU - Xie, Saining

N1 - Publisher Copyright:
Copyright 2020 by the author(s).

PY - 2020

Y1 - 2020

N2 - Neural networks are often represented as graphs of connections between neurons. However, despite their wide use, there is currently little understanding of the relationship between the graph structure of the neural network and its predictive performance. Here we systematically investigate how does the graph structure of neural networks affect their predictive performance. To this end, we develop a novel graph-based representation of neural networks called relational graph, where layers of neural network computation correspond to rounds of message exchange along the graph structure. Using this representation we show that: (1) a “sweet spot” of relational graphs leads to neural networks with significantly improved predictive performance; (2) neural network’s performance is approximately a smooth function of the clustering coefficient and average path length of its relational graph; (3) our findings are consistent across many different tasks and datasets; (4) the sweet spot can be identified efficiently; (5) top-performing neural networks have graph structure surprisingly similar to those of real biological neural networks. Our work opens new directions for the design of neural architectures and the understanding on neural networks in general.

AB - Neural networks are often represented as graphs of connections between neurons. However, despite their wide use, there is currently little understanding of the relationship between the graph structure of the neural network and its predictive performance. Here we systematically investigate how does the graph structure of neural networks affect their predictive performance. To this end, we develop a novel graph-based representation of neural networks called relational graph, where layers of neural network computation correspond to rounds of message exchange along the graph structure. Using this representation we show that: (1) a “sweet spot” of relational graphs leads to neural networks with significantly improved predictive performance; (2) neural network’s performance is approximately a smooth function of the clustering coefficient and average path length of its relational graph; (3) our findings are consistent across many different tasks and datasets; (4) the sweet spot can be identified efficiently; (5) top-performing neural networks have graph structure surprisingly similar to those of real biological neural networks. Our work opens new directions for the design of neural architectures and the understanding on neural networks in general.

UR - http://www.scopus.com/inward/record.url?scp=85101822373&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85101822373&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85101822373

T3 - 37th International Conference on Machine Learning, ICML 2020

SP - 10812

EP - 10822

BT - 37th International Conference on Machine Learning, ICML 2020

A2 - Daume, Hal

A2 - Singh, Aarti

PB - International Machine Learning Society (IMLS)

T2 - 37th International Conference on Machine Learning, ICML 2020

Y2 - 13 July 2020 through 18 July 2020

ER -