Dr. Guang Wang has another paper accepted by IJCAI 2025
Dr. Guang Wang, an Assistant Professor in the Computer Science Department, and his recent research has been accepted by in the the 34th International Joint Conference on Artificial Intelligence (IJCAI 2025) AI and Social Good track. The acceptance rate is 18.5%. The paper title is “Uncertainty-aware Predict-Then-Optimize Framework for Equitable Post-Disaster Power Restoration”, and the first author is Dr. Wang’s current PhD student Lin Jiang.
In this work, they propose an Equity-aware Predict-then-Optimize Power Restoration framework, called EPOPR. The objective is to minimize the total outage duration while ensuring that outage durations across communities satisfy equity criteria. EPOPR consists of two key components: (1) Equity-Conformalized Quantile Regression (ECQR) for uncertainty-aware repair duration prediction. (2) Spatial-Temporal Attentional Soft Actor-Critic (STA-SAC) for optimizing the repair sequence. We comprehensively evaluate EPOPR using the real-world power outage datasets from Tallahassee, Florida. Experimental results
demonstrate that EPOPR can reduce outage duration and decrease inequity among regions.
The paper will be presented at the IJCAI 2025 conference in August.