Graph Neural Network-Guided Design Space Exploration for Power, Performance, and Area Optimization in Advanced VLSI Circuits
Keywords:
Graph neural networks; design space exploration; VLSI physical design; power-performance-area; constrained optimization; surrogate modelingAbstract
Advanced VLSI circuits at sub-10-nm nodes require design space exploration across discrete and continuous decisions whose effects on power, performance, and area are strongly coupled by circuit topology. Conventional surrogate models treat a candidate configuration as a flat feature vector and therefore lose connectivity, hierarchy, and reconvergent-path information that govern timing and power. This manuscript develops a graph neural network-guided design space exploration framework that represents each candidate circuit as a heterogeneous graph, learns message-passing embeddings of cells, nets, and design variables, and is designed to predict power, performance, and area with calibrated uncertainty. The framework combines a multi-task graph neural network surrogate with constrained Bayesian optimization. A normalized scalarization defines exploration directions, while expected improvement is weighted by constraint-feasibility probabilities for timing, power density, and area budgets. The manuscript formalizes graph construction, message-passing and readout operators, a heteroscedastic predictive model, the training objective, and a constrained acquisition rule. It also analyzes how the framework differs from evolutionary search, Gaussian-process optimization, and tree-ensemble surrogates, and it identifies failure modes related to distribution shift, surrogate exploitation, graph scaling, and constraint correlation. The contribution is a technically explicit methodological foundation for topology-aware power-performance-area exploration that keeps trade-offs visible rather than collapsing them prematurely. The framework is intended to guide reproducible implementation and rigorous evaluation in advanced-node design flows.
References
1. Kahng, A. B. (2018). Machine learning applications in physical design: Recent results and directions. In Proceedings of the 2018 International Symposium on Physical Design (pp. 68–73). ACM.
2. Mirhoseini, A., Goldie, A., Yazgan, M., Jiang, J., Songhori, E., Wang, S., Lee, Y., Johnson, E., Pathak, O., Nazi, A., Pak, J., Tong, A., Srinivasa, K., Hang, W., Hui, E., Bansal, A., Cousins, D., Shafi, W., Kaufman, M., ... Dean, J. (2021). A graph placement methodology for fast chip design. Nature, 594(7862), 207–212.
3. Huang, G., Hu, J., He, Y., Liu, J., Ma, M., Shen, Z., Wu, J., Xu, Y., Zhang, H., Zhong, K., Xue, N., Tong, Y., Wang, P., Zhou, Y., Li, Y., & Xie, Y. (2021). Machine learning for electronic design automation: A survey. ACM Transactions on Design Automation of Electronic Systems, 26(5), 1–46.
4. Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021). A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems, 32(1), 4–24.
5. Kipf, T. N., & Welling, M. (2017). Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations.
6. Xu, K., Hu, W., Leskovec, J., & Jegelka, S. (2019). How powerful are graph neural networks? In International Conference on Learning Representations.
7. Li, Y., Tarlow, D., Brockschmidt, M., & Zemel, R. (2016). Gated graph sequence neural networks. In International Conference on Learning Representations.
8. Gilmer, J., Schoenholz, S. S., Riley, P. F., Vinyals, O., & Dahl, G. E. (2017). Neural message passing for quantum chemistry. In Proceedings of the 34th International Conference on Machine Learning (pp. 1263–1272). PMLR.
9. Deb, K., Pratap, A., Agarwal, S., & Meyarivan, T. (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE Transactions on Evolutionary Computation, 6(2), 182–197.
10. Shahriari, B., Swersky, K., Wang, Z., Adams, R. P., & de Freitas, N. (2016). Taking the human out of the loop: A review of Bayesian optimization. Proceedings of the IEEE, 104(1), 148–175.
11. Snoek, J., Larochelle, H., & Adams, R. P. (2012). Practical Bayesian optimization of machine learning algorithms. In Advances in Neural Information Processing Systems (Vol. 25, pp. 2951–2959).
12. Forrester, A. I. J., Sóbester, A., & Keane, A. J. (2008). Engineering design via surrogate modelling: A practical guide. Wiley.
13. Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 785–794). ACM.
14. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
15. Rasmussen, C. E., & Williams, C. K. I. (2006). Gaussian processes for machine learning. MIT Press.
16. Peng, Hongwu, et al. "Rrnet: Towards relu-reduced neural network for two-party computation based private inference." arXiv preprint arXiv:2302.02292 (2023).
17. Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. In International Conference on Learning Representations.
18. Ioffe, S., & Szegedy, C. (2015). Batch normalization: Accelerating deep network training by reducing internal covariate shift. In Proceedings of the 32nd International Conference on Machine Learning (pp. 448–456). PMLR.
19. Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I., & Salakhutdinov, R. (2014). Dropout: A simple way to prevent neural networks from overfitting. Journal of Machine Learning Research, 15(1), 1929–1958.
20. Bishop, C. M. (2006). Pattern recognition and machine learning. Springer.
21. Settles, B. (2009). Active learning literature survey (Computer Sciences Technical Report 1648). University of Wisconsin–Madison.
22. Binois, M., & Wycoff, N. (2022). A survey on high-dimensional Gaussian process modeling with application to Bayesian optimization. ACM Transactions on Evolutionary Learning and Optimization, 2(2), 1–52.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Computational Intelligence Systems

This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.