SHIMIAO LI

Shimiao (Cindy) Li 李诗淼

shimiaol [at] buffalo.edu

I am an assistant professor at the Dept. of EE at University at Buffalo, and an affiliate faculty member at the Institute for Artificial Intelligence and Data Science (IAD). Prior to joining UB, I obtained my PhD from Carnegie Mellon University in 2024, working with Prof. Larry Pileggi. My research interest lies on sparse optimization and AI-driven intelligence for power grid resilience. My work has been recognized by the best paper program in the 2021 & 2026 PES general meeting, and selected in 2023 Microsoft Accelerate Foundation Models Research Initiative.

Main ongoing research:

Sparse optimization for grid resilience
AI-driven intelligence for grid planning (in funded partnership with the New York Power Authority).

I am currently accepting PhD applications. Please email your CV, transcripts, and English test scores to [email protected], if you are interested in joining the group!

News:

💥Jul 2026: Congratulations! Our work "Sparsity-exploiting Gaussian Process for Robust Transient Learning of Power System Dynamics" has been selected for Best Paper Session at PESGM26. (Top 60 papers among a total of 1,369 submissions) See blog for details!
Mar 10 2026: I'm giving a talk at UB IAD Tech Exchange Seminar Series. 12:30-1pm, 2026 Lockwood Hall, UB North Campus.
Mar 2026: Our work "Sparsity-exploiting Gaussian Process for Robust Transient Learning of Power System Dynamics." has been accepted by PESGM 2026. You may find our arxiv preprint arXiv: 2510.15150.
Mar 2026: Our work "Homotopy-Guided Self-Supervised Learning of Parametric Solutions for AC Optimal Power Flow has been accepted by PESGM2026, you may find our arXiv preprint arXiv:2511.11677.
Feb 2026: Our work "Frequency-Aware Sparse Optimization for Diagnosing Grid Instabilities and Collapses." has been accepted by PESGM2026. Congratulations to Swadesh Vhakta. You may find our preprint here arXiv:2511.07553.
Feb 2026: I began serving on the IEEE PES Power System Operation, Planning, and Economic (PSOPE) technical committee as the webmaster.
Jan 2026: Our work "Multi-Period Sparse Optimization for Proactive Grid Blackout Diagnosis.” has been accepted by power system computation conference (PSCC) 2026. Congratulations to Qinghua Ma! Full paper to appear at PSCC2026, but you may find our preprint here: arXiv preprint arXiv:2510.14045.
Dec 2025: I became an affiliated faculty member of Institute for Artificial Intelligence and Data Science (IAD) at University at Buffalo.
Jun 2025: Glad to serve on the NSF review panel for the EPCN program!
Jan 2025: Happy to serve in the TPC for DAC 2025 (Design Automation Conference) in the "AI1. AI/ML Algorithms" track.
💥Jan 2025: Happy to start my new position at University at Buffalo SUNY as an assistant professor!
--------------------------
💥Dec 2024: I obtained my PhD in ECE from Carnegie Mellon University. My PhD thesis is titled "Exploiting sparse structures and synergy designs to advance situational awareness of electrical power grid". See my paper and defense slides here!
Nov 15 2024: I have passed my PhD thesis defense!
Sep 2024: Happy to give a seminar talk titled "Physics-ML Synergy for power grid situation awareness: bridging the gap between system and data" at PES Oregon Chapter Meeting!
July 2024: Happy to present our work "A Convex Method of Generalized State Estimation using Circuit-theoretic Node-breaker Model" in 2024 PES General Meeting at Seattle.
💥I will visit the EE Department at University of Buffalo on March 26th 2024. Looking forward to it!
💥I will visit the CEE (Civil and Environmental Engingeering) Department at CMU on 3/6/2024. Looking forward to it!
💥I will visit the ECE Departmet at UMASS Lowell on 3/1/2024. Looking forward to meeting you there!
💥I will visit the ECE Department at UMASS Amherst on 2/26/2024. Looking forward to it! Feel free to come to my siminar and chat (4pm Marston 132)!
Feb 2024: I will give a seminar talk at CMU, on Thursday February 22, 2024 at 12:00 noon ET, The talk will be offered in person in CMU Wean Hall 3701 (lunch available!) and online via ZOOM. To participate on zoom, please register online.
Feb 2024: Our paper "Contingency Analysis with Warm Starter using Probabilistic Graphical Model." has been accepted by PSCC 2024 – Power Systems Computation Conference and now published on Electric Power Systems Research (EPSR) journal!
Nov 2023: Our paper "A Convex Method of Generalized State Estimation using Circuit-theoretic Node-breaker Model" is accepted by IEEE transactions on powr systems!
Oct 2023: Invited talk in CEIC advisory committee meeting at CMU, about my recent work "Physics-ML synergy toward Maximal Situation Awareness". Some high-level idea and preliminary results are available here.
Sep 2023: My proposed work "Physics-ML Synergy Methodology for Maximal Situation Awareness of Critical Infrastructure" was selected for the Microsoft Accelerate Foundation Models Research Initiative.
Jul 2023: I presented my work "Dynamic Graph-Based Anomaly Detection in the Electrical Grid" in the transaction paper session of the IEEE PES general meeting (July 16-23, Orlando).
Jul 2023: Our work "Combined Transmission and Distribution State-Estimation for Future Electric Grids." is accepted as a chapter in book "Power Systems Operation with 100% Renewable Energy Sources". This is an extension of my state estimation works to combined transmission and distribution networks.
Jun 2023: Oral presentation at 2023 e-energy conference companion workshop (AMLIES) about power grid bahavioral patterns and related risks of generalization in applied ML methods, see paper here.
May 2023: I completed my thesis prospectus! The research topic is "Physics-ML synergy toward maximal situation awareness on power grids".
Dec 2022: Our work "Shedding Light on Inconsistencies in Grid Cybersecurity: Disconnects and Recommendations." is accepted by 2023 IEEE Symposium on Security and Privacy (SP)
Nov 2022: My new work GridWarm is a probabilistic graphical model combined with neural network tools to predict system conditions under cyber-attack events; predictions used to warm start power system contingency analysis and results in 5x faster contingency analysis! see paper here, and GitHub here.
Oct 2022: Oral presentation at CEIC advisory committee meeting, CMU.
Jun-Aug 2022: Internship in the optimization and control group in Pacific Northwest National Laboratory (PNNL). My work is about using homotopy based meta-learning heuristics to improve the training of unsupervised neural networks to better learn constrained optimization problem solutions. This article describes our work: "Homotopy Learning of Parametric Solutions to Constrained Optimization Problems".
May 2022: course project: AugMax for ASR, an adversarial attack on automatic speech recognition (ASR), see Video Presentation, Github
Dec 2021: A MatPower-based cyber-security modelling of AC false data injection attack (FDIA) with more realistic constraints: attackers having inaccurate topology, network parameters, state estimate and limited meter writing access. See GitHub.
Nov 2021: A transaction paper "Dynamic Graph-Based Anomaly Detection in the Electrical Grid" accepted by IEEE transactions on power systems.
Oct 2021: Oral presentation at CEIC advisory committee meeting, CMU.
May 2021: Shimiao Li, Amritanshu Pandey, and Larry Pileggi. "A WLAV-based robust hybrid state estimation using circuit-theoretic approach." won Best Paper Award in PES General Meeting 2021.
Oct 2020: Oral presentation at CEIC advisory committee meeting, CMU.