当前位置: X-MOL 学术Wireless Netw. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A Hyperparameters automatic optimization method of time graph convolution network model for traffic prediction
Wireless Networks ( IF 2.1 ) Pub Date : 2021-06-28 , DOI: 10.1007/s11276-021-02672-5
Lei Chen , Lulu Bei , Yuan An , Kailiang Zhang , Ping Cui

Smart transportation is an essential component of the smart city. Traffic prediction is an important issue in smart transportation. The convolutional neural networks (GCN) are an effective approach for traffic prediction. However, the GCN meets some challenges, such as stability of prediction precision and computation cost, in traffic prediction. The hyperparameters significantly affect the performance of GCN. We conduct a regression analysis between hyperparameters and GCN performance. Our simulation results show that there is the obvious optimal point of hyperparameters. Some empirical suggestion is given to adjust the hyperparameters based on the simulation results.



中文翻译:

一种用于交通预测的时间图卷积网络模型的超参数自动优化方法

智慧交通是智慧城市的重要组成部分。交通预测是智能交通中的一个重要问题。卷积神经网络 (GCN) 是一种有效的交通预测方法。然而,GCN 在流量预测中遇到了一些挑战,例如预测精度的稳定性和计算成本。超参数显着影响 GCN 的性能。我们在超参数和 GCN 性能之间进行回归分析。我们的仿真结果表明,超参数存在明显的最优点。根据仿真结果给出了一些调整超参数的经验建议。

更新日期:2021-06-29
down
wechat
bug