Quantum walk neural networks with feature dependent coins

Recent neural networks designed to operate on graph-structured data have proven effective in many domains. These graph neural networks often diffuse information using the spatial structure of the graph. We propose a quantum walk neural network that learns a diffusion operation that is not only depen...

Full description

Saved in:
Bibliographic Details
Published inApplied network science Vol. 4; no. 1; pp. 1 - 16
Main Authors Dernbach, Stefan, Mohseni-Kabir, Arman, Pal, Siddharth, Gepner, Miles, Towsley, Don
Format Journal Article
LanguageEnglish
Published Cham Springer International Publishing 23.09.2019
Springer Nature B.V
SpringerOpen
Subjects
Cognitive tasks
Coins
Complexity
Computer Appl. in Social and Behavioral Sciences
Computer Science
Graph neural networks
Learning
Neural networks
Quantum random walks
Random walk
Random walk theory
Random walks
Simulation and Modeling
Special Issue of the 7th International Conference on Complex Networks and Their Applications
Random walks
Quantum random walks
Graph neural networks
Online AccessGet full text

Cover

More Information
Summary:Recent neural networks designed to operate on graph-structured data have proven effective in many domains. These graph neural networks often diffuse information using the spatial structure of the graph. We propose a quantum walk neural network that learns a diffusion operation that is not only dependent on the geometry of the graph but also on the features of the nodes and the learning task. A quantum walk neural network is based on learning the coin operators that determine the behavior of quantum random walks, the quantum parallel to classical random walks. We demonstrate the effectiveness of our method on multiple classification and regression tasks at both node and graph levels.
ISSN:2364-8228
2364-8228
DOI:10.1007/s41109-019-0188-2