MDTGAN: Multi domain generative adversarial transfer learning network for traffic data imputation

• Published in: Expert systems with applications 2024-12, Vol.255, p.124478, Article 124478
• Main Authors: Fang, Jie, He, Hangyu, Xu, Mengyun, Chen, Hongting
• Format: Article
• Language: English
• Subjects: Domain discriminator; Generative adversarial network; Traffic data imputation; Transfer learning
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2024.124478

•Introduce multi-domain data to argument training samples, enabling accurate data imputation even with high missing rates.•MDTGAN leverages node-level road data to adapt to diverse topologies, learning generalized spatiotemporal knowledge.•Domain discriminator guides spatial encoder to learn public knowledge for mining domain-invariant features.•MDTGAN outperforms baselines on real datasets, achieving state-of-the-art traffic data imputation performance. Accurate, real-time, and efficient traffic data, crucial for intelligent transportation systems, which is often disrupted in the real world due to the influence of weather, interference, and equipment failures. Generative Adversarial Networks (GANs), have achieved significant results in image restoration, providing insights for traffic data imputation. Recent research has focused on developing more effective and reasonable model by learning transferable common knowledge from different cities or different scenarios, which is also an excellent idea for improving the data imputation performance. In light of, we contrive to design a GANs based transferable traffic data imputation model, namely Multi Domain Generative Adversarial Transfer Learning Network (MDTGAN). Firstly, the model consists of two stages. In pre-training stage, the model utilizes the source domain datasets (Similar cities road network datasets) to update parameters and transferred them. Then, the model parameters are optimized using the target domain dataset (Target city road network dataset) in fine-tuning stage to avoid the issue of insufficient samples caused by data missing. Secondly, to adapt the model to different dataset topologies, MDTGAN models the node-level data by taking node time series as input, enabling the model to autonomously learn the prevalent spatiotemporal correlations inherent in nodes. Additionally, we introduce a domain discriminator module that guides the spatial encoder in learning domain-invariant features of network node spatial information, enhancing the model generalization ability. The experiments conducted on three publicly available datasets, in which one dataset is regarded as the target dataset, while the others serve as source datasets. The experimental results demonstrate that the MDTGAN model consistently outperforms the baseline models. Specifically, the MDTGAN model can transfer valuable knowledge from the source domain datasets to improve data imputation performance on the target domain dataset, provi