Positional Encoding Enhanced Graph Quantile Learning for Uncertainty-Aware Spatial Prediction
DOI:
https://doi.org/10.24237/djes.2026.19304Keywords:
Spatial Data Prediction, Graph Neural Networks, Quantile Regression, Positional Encoding, Uncertainty QuantificationAbstract
Spatial prediction is instrumental in environmental monitoring, urban studies, and geostatistics where accuracy and dependable uncertainty bounds are mandatory. This work introduces the Graph Quantile Neural Network (GQNN) with Positional Encoding (PE), which integrates graph-based spatial representation learning, the sinusoidal positional encoding, and ordered multi-quantile regression. The controlled benchmark is randomized to 70/ 15/ 15 train- validation- test splits and includes 100–1000 irregular nodes with k-nearest-neighbor graph construction (k = 8). Concerning practical relevance, the revised evaluation also specifies out-of-sample tests with public U.S. EPA AirData 2024 annual PM2.5 monitor data and the California Housing geospatial data set, with the same split for GQNN-PE, Ordinary Kriging, and Gaussian Process Regression (GPR). GQNN-PE results in MAE = 0.653, RMSE = 0.972 and determination coefficient (R 2) of 0.89, with PICP = 94.3% and MPIW = 2.11, which provides a decrease of RMSE by 24.3% and 17.0% with respect to Kriging and GPR respectively. For practical relevance, an explicit inverse z-score transformation is also provided so that normalized MAE, RMSE, and interval width can be reported in physical units on real data. Five-seed statistical testing gives p < 0.01.
Downloads
References
[1] N. Cressie, M. Sainsbury-Dale, and A. Zammit-Mangion, “Basis-function models in spatial statistics,” Annu. Rev. Stat. Appl., vol. 9, pp. 373–400, 2022.
[2] K. Klemmer, N. S. Safir, and D. B. Neill, “Positional encoder graph neural networks for geographic data,” in Proc. Int. Conf. Artif. Intell. Statist. (AISTATS), 2023, pp. 1379–1389.
[3] L. Wu, P. Cui, J. Pei, L. Zhao, and X. Guo, “Graph neural networks: Foundations, frontiers and applications,” in Proc. 28th ACM SIGKDD Conf. Knowl. Discovery Data Mining (KDD), 2022, pp. 4840–4841.
[4] F. Rodrigues and F. C. Pereira, “Beyond expectation: Deep joint mean and quantile regression for spatiotemporal problems,” IEEE Trans. Neural Netw. Learn. Syst., vol. 31, no. 12, pp. 5377–5389, 2020.
[5] V. Kuleshov and S. Deshpande, “Calibrated and sharp uncertainties in deep learning via density estimation,” in Proc. Int. Conf. Mach. Learn. (ICML), 2022, pp. 11683–11693.
[6] Y. Wu, D. Zhuang, A. Labbé, and L. Sun, “Inductive graph neural networks for spatiotemporal kriging,” in Proc. AAAI Conf. Artif. Intell., vol. 35, no. 5, pp. 4478–4485, 2021, doi: 10.1609/aaai.v35i5.16575.
[7] D. Zhuang, Y. Bu, G. Wang, S. Wang, and J. Zhao, “SAUC: Sparsity-aware uncertainty calibration for spatiotemporal prediction with graph neural networks,” in Proc. 32nd ACM Int. Conf. Adv. Geographic Inf. Syst. (SIGSPATIAL), Atlanta, GA, USA, 2024, pp. 160–172, doi: 10.1145/3678717.3691241
[8] L. Poggio, L. M. De Sousa, N. H. Batjes, G. B. Heuvelink, B. Kempen, E. Ribeiro, and D. Rossiter, “SoilGrids 2.0: Producing soil information for the globe with quantified spatial uncertainty,” SOIL, vol. 7, no. 1, pp. 217–240, 2021, doi: 10.5194/soil-7-217-2021.
[9] H. Mao, R. Martin, and B. J. Reich, “Valid model-free spatial prediction,” J. Amer. Stat. Assoc., vol. 119, no. 546, pp. 904–914, 2024, doi: 10.1080/01621459.2022.2147531.
[10] M. Zhang, D. Yu, Y. Li, and L. Zhao, “Deep geometric neural network for spatial interpolation,” in Proc. 30th Int. Conf. Adv. Geographic Inf. Syst. (SIGSPATIAL), Seattle, WA, USA, 2022, pp. 1–4.
[11] J. Cheng, K. Li, Y. Liang, L. Sun, J. Yan, and Y. Wu, “Rethinking urban mobility prediction: A multivariate time series forecasting approach,” IEEE Trans. Intell. Transp. Syst., 2025.
[12] J. Han, W. Zhang, H. Liu, T. Tao, N. Tan, and H. Xiong, “BigST: Linear-complexity spatio-temporal graph neural network for traffic forecasting on large-scale road networks,” Proc. VLDB Endowment, vol. 17, no. 5, pp. 1081–1090, 2024.
[13] J. Feng, Z. Yang, F. Xu, H. Yu, M. Wang, and Y. Li, “Learning to simulate human mobility,” in Proc. 26th ACM SIGKDD Int. Conf. Knowl. Discovery Data Mining (KDD), 2020, pp. 3426–3433.
[14] Y. Guo, Z. Wang, J. Xue, and Z. Shao, “A spatio-temporal series data model with efficient indexing and layout for cloud-based trajectory data management,” in Proc. IEEE 40th Int. Conf. Data Eng. (ICDE), 2024, pp. 1171–1184.
[15] Z. Fang, L. Pan, L. Chen, Y. Du, and Y. Gao, “MDTP: A multi-source deep traffic prediction framework over spatio-temporal trajectory data,” Proc. VLDB Endowment, vol. 14, no. 8, pp. 1289–1297, 2021.
[16] S. J. Fox, M. Lachmann, M. Tec, R. Pasco, S. Woody, Z. Du, X. Wang, T. A. Ingle, E. Javan, and M. Dahan, “Real-time pandemic surveillance using hospital admissions and mobility data,” Proc. Natl. Acad. Sci. USA, vol. 119, no. 7, Art. no. e2111870119, 2022.
[17] J. Hu, X. Liu, Z. Fan, Y. Liang, and R. Zimmermann, “Towards unifying diffusion models for probabilistic spatio-temporal graph learning,” in Proc. 32nd ACM Int. Conf. Adv. Geographic Inf. Syst. (SIGSPATIAL), Atlanta, GA, USA, 2024, pp. 135–146, doi: 10.1145/3678717.3691235.
[18] K. Bi, L. Xie, H. Zhang, X. Chen, X. Gu, and Q. Tian, “Accurate medium-range global weather forecasting with 3D neural networks,” Nature, vol. 619, no. 7970, pp. 533–538, 2023.
[19] Y.-T. Chou, W.-T. Chang, J. G. Jean, K.-H. Chang, Y.-N. Huang, and C.-S. Chen, “StructGNN: An efficient graph neural network framework for static structural analysis,” Comput. Struct., vol. 299, Art. no. 107385, 2024, doi: 10.1016/j.compstruc.2024.107385.
[20] X. Ouyang, Y. Li, D. Guo, W. Huang, X. Yang, Y. Yang, J. Zhang, and T. Li, “Enhancing few-sample spatio-temporal prediction via relational fusion-based hypergraph neural network,” Inf. Fusion, vol. 121, Art. no. 103149, 2025, doi: 10.1016/j.inffus.2025.103149.
[21] A. Tang, C. Li, J. Yang, H. Zhang, Q. Zheng, and J. Zhang, “Training and application of graph neural networks for predicting structural responses targeted at tall building structures,” J. Build. Eng., vol. 103, Art. no. 112131, 2025, doi: 10.1016/j.jobe.2025.112131.
[22] A. Derrow-Pinion, J. She, D. Wong, O. Lange, T. Hester, L. Perez, M. Nunkesser, S. Lee, X. Guo, and B. Wiltshire, “ETA prediction with graph neural networks in Google Maps,” in Proc. 30th ACM Int. Conf. Inf. Knowl. Manage. (CIKM), 2021, pp. 3767–3776.
[23] S. Xu, X. Fu, J. Cao, B. Liu, and Z. Wang, “Survey on user location prediction based on geo-social networking data,” World Wide Web, vol. 23, no. 3, pp. 1621–1664, 2020.
[24] P. Luo, C. Chen, S. Gao, X. Zhang, D. M. Chol, Z. Yang, and L. Meng, “Understanding the predictability and uncertainty in population distributions empowered by visual analytics,” Int. J. Geogr. Inf. Sci., vol. 39, no. 3, pp. 675–705, 2025, doi: 10.1080/13658816.2024.2427870.
[25] A. Chan, A. Alaa, Z. Qian, and M. Van Der Schaar, “Unlabelled data improves Bayesian uncertainty calibration under covariate shift,” in Proc. Int. Conf. Mach. Learn. (ICML), 2020, pp. 1392–1402.
[26] G. Liu, X. Zhou, Q. Li, Y. Shi, G. Guo, L. Zhao, J. Wang, Y. Su, and C. Zhang, “Spatial distribution prediction of soil As in a large-scale arsenic slag contaminated site based on an integrated model and multi-source environmental data,” Environ. Pollut., vol. 267, Art. no. 115631, 2020, doi: 10.1016/j.envpol.2020.115631.
[27] M. Sachdeva, A. S. Fotheringham, Z. Li, and H. Yu, “On the local modeling of count data: Multiscale geographically weighted Poisson regression,” Int. J. Geogr. Inf. Sci., vol. 37, no. 10, pp. 2238–2261, 2023, doi: 10.1080/13658816.2023.2250838.
[28] Z. Al Mokhtar and S. Dawwd, “3D VAE video prediction model with Kullback–Leibler loss enhancement,” Int. Arab J. Inf. Technol., vol. 21, no. 5, pp. 879–888, 2024, doi: 10.34028/iajit/21/5/9.
[29] U.S. Environmental Protection Agency, “AirData: Air Quality Data Collected at Outdoor Monitors Across the US,” Annual Concentration by Monitor, 2024. [Online]. Available: https://aqs.epa.gov/aqsweb/airdata/download_files.html. Accessed: Aug. 10, 2026.
[30] R. K. Pace and R. Barry, “Sparse spatial autoregressions,” Stat. Probab. Lett., vol. 33, no. 3, pp. 291–297, 1997.
[31] Scikit-learn Developers, “California housing dataset,” scikit-learn documentation. [Online]. Available: https://scikit-learn.org/stable/modules/generated/sklearn.datasets.fetch_california_housing.html. Accessed: Aug. 10, 2026.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Srinivasu Badugu, R. Manivannan, D Radhika
This work is licensed under a Creative Commons Attribution 4.0 International License.
