Talks at INFORMS Annual Meeting

I was invited to present two of our recent works on learning to solve optimization, with applications in network flow and optimal power flow problems. More details about the two virtual talks, “Learning Generalizable Network Flow Solvers via Neural Networks” and “Learning To Solve Optimal Power Flow Via Robust Neural Decoding” can be viewed here!

I am also serving as a FacilitatORs in two sessions with topics on Robust Optimization and Machine Learning respectively. Look forward to virtual INFORMS this year!

Paper Duo on Machine learning for Solving OPF

Two of our papers are on Arxiv trying to use machine learning for efficiently solving Optimal power flow (OPF) problem.

In our first paper, we proposed a novel approach for learning the “codewords”  associated with binding constraints in the OPF problem. Such learned properties can help accelerate OPF solving process.

In the second paper, we theoretically analyze the generalization properties of the learning algorithm to unseen patterns, while deriving novel learning procedures to make the solutions feasible.

Work on Helping to Improve Load Forecasting During COVID-19

We recently posted a preprint Using Mobility for Load Forecasting During the COVID-19 Pandemic on Arxiv. We proposed a novel load forecasting method to use the mobility data and transfer learning to help forecast the volatile load during the pandemic. We hope such method would ease the burden for utilities and system operators during the difficult times. Check out the open-source code and dataset here!

 

Also featured in VentureBeat.

INFORMS 2019

Felt great to attend a local INFORMS annual meeting in Seattle last week! Got to know some very interesting and new research in optimization, energy systems and learning, and got to meet with some old friends.

E-Energy Best Paper Runner-Up

Great honor to get the 2nd Best Paper award at this year’s E-Energy Conference in Phoenix! Trying to advocate more rigorous analysis when applying machine learning/forecasting techniques in power systems. See our paper here!

Also great to attend FCRC 2019! Learned a lot from other venues other than e-Energy such as Sigmetrics, EC and COLT.