Uncertainty-aware Predict-Then-Optimize Framework for Equitable Post-Disaster Power Restoration (Nov 4)

Speaker: Lin Jiang

Date: Nov 4, 11:45 AM – 12: 45 PM

Abstract: The increasing frequency of extreme weather events, such as hurricanes, highlights the gent need for efficient and equitable power system restoration. Many electricity providers make restoration decisions primarily based on the volume of power restoration requests from each region. However, our data-driven analysis reveals significant disparities in request submission volume, as disadvantaged communities tend to submit fewer restoration requests. This disparity makes the current restoration solution inequitable, reaving these communities vulnerable to extended power outages. To address this, we aim to propose an equity-aware power restoration strategy that balances both restoration efficiency and equity across communities. However, achieving this goal is challenging for two reasons: the difficulty of predicting repair durations under dataset heteroscedasticity, and the tendency of reinforcement learning agents to favor low-uncertainty actions, which potentially undermine equity. To overcome these challenges, we design a predict-then-optimize framework called EPOPR with two key components: (1) Equity-Conformalized Quantile Regression for uncertainty-aware repair duration prediction, and (2) Spatial-Temporal Attentional RL that adapts to varying uncertainty levels across regions for equitable decision-making. Experimental results show that our EPOPR effectively reduces the average power outage duration by 3.60% and decreases inequity between different communities by 14.19% compared to state-of-the-art baselines.

Speaker Profile: Lin Jiang is a third-year Ph.D. student in the Department of Computer Science at Florida State University, advised by Dr. Guang Wang in the Data, Computing, and Society (DCS) Lab. His research focuses on trustworthy decision-making, generative AI, and spatial-temporal data mining. Prior to joining FSU, Lin received both his Bachelor’s and Master’s degrees from Southeast University. His work has been published in top conferences and journals in AI, data mining, and mobile computing, including KDD, IJCAI, SIGSPATIAL, and TMC. He is also a travel grant awardee of IJCAI 2025 and KDD 2023. Lin actively serves as a reviewer for major AI conferences, including ICLR, IJCAI, AAAI, SIGKDD, and WWW. He is passionate about discovering meaningful data patterns from real-world datasets and applying AI methods to solve practical problems in decision-making and generative modeling.

Location: LOV 307 (In Person Only)