Dr. Shiyu Liu obtained his PhD in Engineering from the University of Electronic Science and Technology of China in July 2024. He is currently an assistant researcher at the Institute for Statistics and Interdisciplinary Innovation, Southwestern University of Finance and Economics. His primary research focuses on statistical learning, graph neural networks, and federated learning.
Shiyu Liu
/ Assistant Research Fellow
E-mail：liushiyu AT swufe.edu.cn
Awards and Honors
Research Projects
- 2025.01-2025.12: Youth Teacher Growth Project of the Fundamental Research Funds for the Central Universities, “Research on trustworthy modeling methods and theories for privacy-sensitive data”.
Selected Papers
- Zeng, D., Xu, Z., Liu, S., Pan, Y., Wang, Q., & Tang, X. (2025). On the power of adaptive weighted aggregation in heterogeneous federated learning and beyond. International Conference on Artificial Intelligence and Statistics (AISTATS) .
- Luo, J., Chen, G., Zhang, Y., Liu, S., Wang, H., Yu, Y., … & Xu, Z. (2025). Centaur: bridging the impossible trinity of privacy, efficiency, and performance in privacy-preserving transformer inference. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL), 22751-22770.
- Liu, S., Shi, W., Lv, S., Zhang, Y., Wang, H., & Xu, Z. (2024). Meta-learning via PAC-Bayesian with data-dependent prior: generalization bounds from local entropy. International Joint Conferences on Artificial Intelligence (IJCAI).
- Liu, S., Luo, J., Zhang, Y., Wang, H., Yu, Y., & Xu, Z. (2024). Efficient privacy-preserving Gaussian process via secure multi-party computation. Journal of Systems Architecture, 151, 103134.
- Liu, S., Wei, L., Lv, S., & Li, M. (2023). Stability and generalization of lp-regularized stochastic learning for graph convolutional networks. International Joint Conferences on Artificial Intelligence (IJCAI).
- Ai, Q., Liu, S., He, L., & Xu, Z. (2023). Stein variational gradient descent with multiple kernels. Cognitive Computation, 15, 672-682.
- Ai, Q., He, L., Liu, S., & Xu, Z. (2022). ByPE-VAE: Bayesian pseudocoresets exemplar VAE. Neural Information Processing Systems (NeurIPS).