I am Xueying Liu, currently a fifth-year Ph.D. student in the Department of Statistics at Virginia Tech (VT), under the supervision of Professor Xinwei Deng. I am also a member of VT Statistics and Artificial Intelligence Laboratory (VT-SAIL).

My research focuses on developing advanced statistical and machine learning methods tailored to complex, real-world applications through collaborations with various disciplines. Specifically, my study covers the following topics:

• Multi-task learning for clinical trial data (Bristol Myers Squibb)
• Statistics-guided continual learning for AI tasks (AI Center of Excellence, Deloitte)
• Statistical and network modeling with application to social and behavioral science (Biocomplexity Institute at University of Virginia)
• Modeling complex data and uncertainty quantification

Education

• Ph.D. in Statistics, Virginia Tech, 2020-Present
• M.S. in Statistics, Georgia Tech, 2018-2020
• B.S. in Mathematics, Shanghai University, China 2014-2018

Publications

Peer-Reviewed Journal Articles

1. Liu, X., Hu, Z., Deng, X., and Kuhlman, C. J. (2023). Uncertainty visualization for characterizing heterogeneous human behaviors in discrete dynamical system models, Advances in Complex Systems, 26(3), pp. 2340001-1.
2. Hu, Z., Liu, X., Deng, X., and Kuhlman, C. J. (2024). An uncertainty quantification framework for agent-based modeling and simulation in networked anagram games, Journal of Simulation, 1-19.
3. Chu, S., Liu, X., Marathe, A., and Deng, X. (2024). A latent process approach to change-point detection of mixed-type observations, Quality Engineering, 36(2), 407-426.

Peer-Reviewed Conference Paper

1. Liu, X., Hu, Z., Deng, X., and Kuhlman, C. J. (2022). A Bayesian uncertainty quantification approach for agent-based modeling of networked anagram games, Proceedings of the 2022 Winter Simulation Conference (WSC 2022), pp. 310-321. IEEE.
2. Liu, X., Hu, Z., Deng, X., and Kuhlman, C. J. (2022). Bayesian approach to uncertainty visualization of heterogeneous behaviors in modeling networked anagram games, Proceedings of the 2022 International Conference on Complex Networks and Their Applications (CNA 2022), pp. 595-607.
3. Liu, X., Hu, Z., Deng, X., and Kuhlman, C. J. (2023). A calibration model for bot-like behaviors in agent-based anagram game simulation, Proceedings of the 2023 Winter Simulation Conference (WSC 2023), pp.221-232. IEEE.
4. Liu, X., Hu, Z., Deng, X., and Kuhlman, C. (2023). Learning common knowledge networks via exponential random graph models, Proceedings of the 2023 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM 2023). (acceptance rate 17%)
5. He, H., Liu, X., Kuhlman, C. and Deng, X. (2024). A framework of digital twins for modeling human-subject word formation experiments, Accepted by 2024 Winter Simulation Conference (WSC 2024).
6. He, H., Liu, X., et al. (2024). A successive analysis of online networked common knowledge experiments. Accepted by 2024 Advances in Social Networks Analysis and Mining (ASONAM 2024).
7. Lian, J., Liu, X., et al. (2024). Data composition for continual learning in the application of cyberattack detection. Accepted by 2024 Advances in Social Networks Analysis and Mining (ASONAM 2024).

Manuscripts

1. Liu, X., Jiang, H., Zhang J., and Deng, X. (2024). A multi-task learning approach for identifying prognostic variables across multiple historical clinical trials.
2. Liu, X., Hu, Z., Deng, X., and Kuhlman, C. J. (2024). Learning common knowledge networks via exponential random graph models.

Experience

• Biostatistician Intern, Bristol Myers Squibb, Summit, NJ, USA. 2023.05-2023.08
• Research Assistant, Virginia Tech, Blacksburg, VA, USA. 2021.08-present
• Teaching Assistant, Virginia Tech, Blacksburg, VA, USA. 2020.08-2024.05
• Data Analyst Intern, Nielsen, Shanghai, China. 2017.12-2018.03

Talks

• “Data Composition: A New Experimental Design Problem with Application in Continual Learning”, Invited talk at International Conference on Advances in Interdisciplinary Statistics and Combinatorics (AISC), Greensboro, NC, Oct 2024.
• “A multi-task learning approach for identifying prognostic variables across multiple historical clinical trials”, Talk at 2024 Joint Statistical Meetings (JSM), Portland, OR, Aug 2024.
• “A multi-task learning approach for identifying prognostic variables across multiple historical clinical trials”, Poster presentation at ENAR 2024 Spring Meeting, Baltimore, MD, Mar 2024.
• “A multi-task learning approach for identifying prognostic variables across multiple historical clinical trials”, Poster presentation at Women in Data Science Conference, Blacksburg, VA, Feb 2024.
• “A calibration model for bot-like behaviors in agent-based anagram game simulation”, Talk at 2023 Winter Simulation Conference (WSC), San Antonio, TX, Dec 2023.
• “Learning common knowledge networks via exponential random graph models”, Talk at 2023 Advances in Social Networks Analysis and Mining (ASONAM), Kusadasi, Turkey, Nov 2023.
• “Learning common knowledge networks via exponential random graph models”, Poster presentation at 2023 ASA/IMS Spring Research Conference (SRC), Banff, Canada, May 2023.
• “A Bayesian uncertainty quantification approach for agent-based modeling of networked anagram games”, Talk at 2022 Winter Simulation Conference (WSC), Virtual, Dec 2022.
• “Bayesian approach to uncertainty visualization of heterogeneous behaviors in modeling networked anagram games”, Talk at 2022 Complex Networks and Application (CNA), Virtual, Nov 2022.

Misc

Outside of work, I enjoy dancing, including jazz, hip hop, kpop, and choreography, and I am a member of Chaoxic Dance Crew at Virginia Tech. I also adopted two sibling cats (brother and sister) who have never been separated since birth.