DESIRES Lab @ Peking University

The study of saddle-point problems from a dynamical system perspective and how stability analysis systematically inspires algorithm design

The fundamental limit of how fast an electricity market can be operated without jeopardizing system reliability

We are hiring! Postdoc openings on optimization and learning are available! See some of the details here.

DESIRES Lab is constantly looking for motivated graduate students and postdocs to work on power systems, electricity markets, electric vehicles, online platforms, reinforcement learning, etc. We welcome candidates from diverse backgrounds to apply, including but not limited to power & energy, economics, optimization, control, operations research, algorithms, applied mathematics, statistics, and learning. Feel free to email for more details, preferably with CV, if interested.

We have the university fellowships available for postdocs that provide more benefits. Candidates are encouraged to apply. Check here.

[2025.01] We had three papers accepted to AISTATS, ICLR, and ACM HSCC three days in a row!

[2024.12] Zijian was awarded a grant by China National University Student Innovation & Entrepreneurship Development Program to work on Digital Management System for Anti-Counterfeiting and Traceability of Agricultural Products!

[2024.12] Zhuo was among 20 “Students of the Year Finalist” at PKU!

[2024.11] Prof. Wei Qi from Tsinghua IE visited us and gave a talk on Retail on Autonomous Wheels: Spatial Queueing and Online Facility Location.

[2024.09] Yixuan and Yixuan (two Ph.D. students with the same English first name) joined DESIRES Lab.

[2024.09] Yingzhu received Student Travel Grant and Workshop Support from CDC 2024!

[2024.07] Our paper “Accelerated Saddle Flow Dynamics for Bilinearly Coupled Minimax Problems” was accepted to IEEE CDC 2024.

[2024.07] Prof. Bin Liu from University of Strathclyde gave a talk on Data-Driven Maintenance Optimization in our COOL Research Seminar.

[2024.06] Lihuan Zhang from 01.AI gave a talk on LLM Training Acceleration Techniques & Challenges to share latest LLM advances in industry.

[2024.05] Our paper “Dissipative Gradient Descent Ascent Method: A Control Theory Inspired Algorithm for Min-max Optimization” was accepted to IEEE Control Systems Letters (L-CSS).

[2024.05] Prof. Florian Dorfler from ETH Zurich gave a talk on Data-Enabled Predictive Control (DeePC) in our COOL Research Seminar.

[2024.05] Our extended abstract “Reserve or Not? Ordering Emergency Supplies from an Opportunistic Supplier Facing Yield and Demand Uncertainties” was accepted to INFORMS MSOM Conference.

[2024.05] Our paper “Online Electricity Purchase for Data Center with Dynamic Virtual Battery from Flexibility Aggregation” was accepted to Youth Academic Annual Conference of Chinese Association of Automation (YAC).

[2024.04] Christopher Yeh from Caltech gave a talk on Learning for Robust Decision Making in Energy Systems in our COOL Research Seminar.

[2024.04] Chengrui received the prestigious Summer Undergraduate Research Fellowship (SURF) from Caltech to fund his 10-week stay in Southern California!

[2024.03] Prof. Dawei Shi was invited to give a talk on Data-Enabled Event-Triggered Learning and Control.

[2024.01] We were awarded two grants to study trustworthy AI for power systems and architecture redesign for electricity markets, respectively.

[2024.01] We organized an invited session on Towards Next-Gen Power and Energy Systems in Conference on Fully Actuated System Theory and Applications (Shenzhen, China, May 10 – 12), featuring a dozen of exciting talks from academia and industry. Join us!

More…