James Anderson

Associate Professor, Department of Electrical Engineering Member of the Data Science Institute Fu Foundation School of Engineering and Applied Science [email protected] Office: 709 CEPSR Positions: I consider postdoc applications made through the Columbia Data Science Institute

Bio

I'm an associate professor in the Department of Electrical Engineering at Columbia University, where I am also a member of the Data Science Institute. From 2016 to 2019 I was a senior postdoctoral scholar in the Computing + Mathematical Sciences department at the California Institute of Technology. Prior to Caltech, I held a Junior Research Fellowship at St John's College at the University Oxford and was also affiliated with the Department of Engineering Science. I was awarded my DPhil (PhD) from Oxford in 2012 and the BSc and MSc degrees from the University of Reading in 2005 and 2006 respectively.

Research Interests

robust & distributed control, convex optimization, learning for control, power networks & smart grids

Group Meetings

My students are far more organized than I am and they keep an online group meeting schedule. Please check out the website that Kasra Fallah put together for speaker details. If you are affiliated with Columbia and would like to attend, drop me an email.

Recent Updates

• I received an NSF grant (in collaboration with Brian Plancher at Dartmouth) to study ‘‘Safe Multi-Task Learning and Control for Autonomy at the Edge’’

• Congratulations to Yiqian for defending her PhD thesis “Strategic and Resilient Energy Storage Operations: Economics, Cybersecurity, and Risk Management in Modern Power Grids”. Great work Yiqian!

• Leo wins the CDC outstanding student paper award for our paper ‘‘Learning Stabilizing Policies via an Unstable Subspace Representation’’ - great work Leo! photos

• I'm teaching a new class that I created this Fall - E4620 - Numerical Methods for Data Analysis

• Columbia is hosting the 2nd Northeast Systems and Control Workshop on May 3rd, see you there!

• Our paper on ‘‘Distributed and Localized Model Predictive Control’’ won the IEEE Transactions on Control of Network Systems outstanding paper award – joint work with Carmen, Lisa, and Nik – Columbia story

• Han successfully defended her PhD thesis “Federated Learning for Reinforcement Learning and Control” Congratulations Han!

• I gave a talk on Multi-Task Learning for Control with MAML-LQR at the Johns Hopkins ECE department: slides

• I gave a talk on Collaborative Learning for Control of Heterogeneous Systems as part of the Autonomy Talks virtual seminar series: video

• I gave a talk on Energy Storage Market Power and Strategic Withholding at the NREL Workshop on Autonomous Energy Systems: slides

• We won the best paper award at L4DC for our work on meta-learning for control: photos, Columbia story

• Our work on ‘‘Sample-efficient Rrepresentation Learning from Non-IID Non-isotropic Data’’ was awarded a spotlight presentation at ICLR

• I gave an MIT LIDS seminar titled To Collaborate, or Not? Federated Learning Meet Control

A nearly successful effort to introduce the atmospheric system of traction. Original image: W. Heath Robinson Colouring: Adam Hart-Davis (click on the image to enlarge)