Efficient graph neural architecture search using Monte Carlo Tree search and prediction network

作者：

Highlights：

• Narrow down the search space by a variant of Mont Carlo Tree Search.

• Accelerate the evaluation of architectures by the prediction model.

• Make the training of the agent more efficient and stable.

摘要

•Narrow down the search space by a variant of Mont Carlo Tree Search.•Accelerate the evaluation of architectures by the prediction model.•Make the training of the agent more efficient and stable.

论文关键词：Graph neural architecture search,Graph Neural Networks,Monte Carlo Tree search,Prediction network,Reinforcement learning

论文评审过程：Received 23 October 2021, Revised 9 August 2022, Accepted 24 September 2022, Available online 4 October 2022, Version of Record 20 October 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118916