Adaptive GPU-Accelerated Graph Learning for Systemic Risk Assessment in Interconnected Financial Markets

Authors

  • Henrietta Emilsson Department of Computer Science, University of New Hampshire, Durham, NH, USA.
  • Anita Sellers Department of Computer Science and Engineering, University at Buffalo, Buffalo, NY, USA.
  • Amadeo Nievo Department of Electrical Engineering and Computer Science, University of Missouri, Columbia, MO, USA.
  • Indira Toor Department of Computer Science, Binghamton University, Binghamton, NY, USA.

Keywords:

systemic risk; graph neural networks; GPU acceleration; financial networks; uncertainty calibration; dynamic graph learning

Abstract

Systemic risk in interconnected financial markets is shaped by time-varying exposures, common holdings, funding dependence, and correlated stress that propagate across institutions and market segments. Conventional connectedness measures and network risk models capture important channels of contagion, but they typically rely on static or low-frequency network representations and struggle to combine heterogeneous relational information with scalable inference. This paper develops a methodological framework for adaptive GPU-accelerated graph learning that treats financial markets as dynamic, typed, and weighted graphs and estimates node-level distress and system-level loss through temporal message passing. The central contribution is an integrated design that couples dynamic graph construction, localized graph convolution with recurrent memory, risk-sensitive learning objectives, covariance-based exposure features, and calibrated predictive uncertainty. The framework specifies how GPU execution can support repeated sparse message passing, adaptive neighborhood sampling, and workload-balanced edge processing while preserving the economic semantics of financial linkages. Analytical assessment focuses on the learning objective, propagation operator, computational complexity, calibration logic, and domain-specific failure modes. The design clarifies when graph learning adds value over econometric and network-based benchmarks, how uncertainty and explainability can be incorporated without sacrificing scalability, and which implementation conditions are necessary for supervisory and risk-management use. The resulting contribution is a rigorous computational and analytical foundation for systemic risk assessment rather than a claim of empirical superiority.

References

1. Billio, M., Getmansky, M., Lo, A. W., & Pelizzon, L. (2012). Econometric measures of connectedness and systemic risk in the finance and insurance sectors. Journal of Financial Economics, 104(3), 535–559.

2. Diebold, F. X., & Yılmaz, K. (2014). On the network topology of variance decompositions: Measuring the connectedness of financial firms. Journal of Econometrics, 182(1), 119–134.

3. Acemoglu, D., Ozdaglar, A., & Tahbaz-Salehi, A. (2015). Systemic risk and stability in financial networks. American Economic Review, 105(2), 564–608.

4. Battiston, S., Puliga, M., Kaushik, R., Tasca, P., & Caldarelli, G. (2012). DebtRank: Too central to fail? Financial networks, the FED and systemic risk. Scientific Reports, 2, 541.

5. Haldane, A. G., & May, R. M. (2011). Systemic risk in banking ecosystems. Nature, 469(7330), 351–355.

6. Caccioli, F., Barucca, P., & Kobayashi, T. (2018). Network models of financial systemic risk: A review. Journal of Computational Social Science, 1(1), 81–114.

7. Barabási, A.-L., & Albert, R. (1999). Emergence of scaling in random networks. Science, 286(5439), 509–512.

8. Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations.

9. Veličković, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., & Bengio, Y. (2018). Graph attention networks. In International Conference on Learning Representations.

10. Hamilton, W. L., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. In Advances in Neural Information Processing Systems (Vol. 30, pp. 1024–1034).

11. Scarselli, F., Gori, M., Tsoi, A. C., Hagenbuchner, M., & Monfardini, G. (2009). The graph neural network model. IEEE Transactions on Neural Networks, 20(1), 61–80.

12. Xie, X., Peng, H., Hasan, A., Huang, S., Zhao, J., Fang, H., ... & Ding, C. (2023, October). Accel-gcn: High-performance gpu accelerator design for graph convolution networks. In 2023 IEEE/ACM International Conference on Computer Aided Design (ICCAD) (pp. 01-09). IEEE.

13. Defferrard, M., Bresson, X., & Vandergheynst, P. (2016). Convolutional neural networks on graphs with fast localized spectral filtering. In Advances in Neural Information Processing Systems (Vol. 29, pp. 3844–3852).

14. Zhou, J., Cui, G., Hu, S., Zhang, Z., Yang, C., Liu, Z., Wang, L., Li, C., & Sun, M. (2020). Graph neural networks: A review of methods and applications. AI Open, 1, 57–81.

15. Xu, K., Hu, W., Leskovec, J., & Jegelka, S. (2019). How powerful are graph neural networks? In International Conference on Learning Representations.

16. Gal, Y., & Ghahramani, Z. (2016). Dropout as a Bayesian approximation: Representing model uncertainty in deep learning. In Proceedings of the 33rd International Conference on Machine Learning (pp. 1050–1059).

17. Kendall, A., & Gal, Y. (2017). What uncertainties do we need in Bayesian deep learning for computer vision? In Advances in Neural Information Processing Systems (Vol. 30, pp. 5574–5584).

18. Engle, R. (2002). Dynamic conditional correlation: A simple class of multivariate generalized autoregressive conditional heteroskedasticity models. Journal of Business & Economic Statistics, 20(3), 339–350.

19. Ledoit, O., & Wolf, M. (2004). A well-conditioned estimator for large-dimensional covariance matrices. Journal of Multivariate Analysis, 88(2), 365–411.

20. Battiston, S., Caldarelli, G., May, R. M., Roukny, T., & Stiglitz, J. E. (2016). The price of complexity in financial networks. Proceedings of the National Academy of Sciences, 113(36), 10031–10036.

21. Ying, Z., Bourgeois, D., You, J., Zitnik, M., & Leskovec, J. (2019). GNNExplainer: Generating explanations for graph neural networks. In Advances in Neural Information Processing Systems (Vol. 32, pp. 9240–9251).

22. Sundararajan, M., Taly, A., & Yan, Q. (2017). Axiomatic attribution for deep networks. In Proceedings of the 34th International Conference on Machine Learning (pp. 3319–3328).

Downloads

Published

2026-09-07

How to Cite

Henrietta Emilsson, Anita Sellers, Amadeo Nievo, & Indira Toor. (2026). Adaptive GPU-Accelerated Graph Learning for Systemic Risk Assessment in Interconnected Financial Markets. Global Financial Analytics Research Review, 3(4). Retrieved from https://www.gfarr.org/index.php/home/article/view/171