基于转导图自编码器的保形负荷预测
Conformal Load Prediction with Transductive Graph Autoencoders
- City University of Hong Kong(香港城市大学)
- Royal Holloway, University of London(伦敦大学皇家霍洛威学院)
机构由 AI 辅助整理,请以论文原文为准。
AI总结:
本文提出一种基于图神经网络和保形预测的边权重预测方法,结合误差重加权与保形分位数回归处理异方差性,在交通数据集上实现了优于基线的覆盖率和效率。
AI中文摘要:
预测图上的边权重在交通系统到社交网络等多个领域具有广泛应用。本文提出了一种具有覆盖保证的图神经网络(GNN)边权重预测方法。我们利用保形预测来校准GNN输出,并生成有效的预测区间。通过误差重加权和保形分位数回归(CQR)来处理数据异方差性。我们在真实交通数据集上将所提方法与基线技术进行了性能比较。实验结果表明,我们的方法在覆盖率和效率上均优于所有基线方法,展现出良好的鲁棒性和适应性。
英文摘要:
Predicting edge weights on graphs has various applications, from transportation systems to social networks. This paper describes a Graph Neural Network (GNN) approach for edge weight prediction with guaranteed coverage. We leverage conformal prediction to calibrate the GNN outputs and produce valid prediction intervals. We handle data heteroscedasticity through error reweighting and Conformalized Quantile Regression (CQR). We compare the performance of our method against baseline techniques on real-world transportation datasets. Our approach has better coverage and efficiency than all baselines and showcases robustness and adaptability.