IEOR - Designing a More Efficient World

Postdoc Profile: Richard Zhang

Richard Zhang is currently a postdoctoral scholar in the IEOR department. He has just accepted a faculty position at the University of Illinois, Urbana-Champaign. Below, Richard reflects on his experience at UC Berkeley and talks more about his plans for the future.

Richard Zhang is a postdoctoral scholar in the IEOR department. He has just accepted a faculty position at the University of Illinois Urbana-Champaign.

Tell us a bit about your research. What topics are you interested in? How are you applying IEOR methods to solve them?

My research is on numerical algorithms for computational problems, with applications in energy and transportation. The methods of IEOR allow us to classify a computational problem as being “easier” or “easy” in a principled, rigorous sense. Fortunately, we have been able to classify many important societal problems as “easy” in this manner, including the optimal power flow problem in energy, and the offset optimization problem in transportation.

Tell us a bit about your work here at UC Berkeley with Professor Lavaei. How has his advice helped you succeed?

I joined Professor Lavaei’s group as a specialist in convex optimization. However, it was Professor Lavaei who suggested that I instead study algorithms for nonconvex optimization. In principle, these algorithms can become stuck at a local minimum. However, this never seemed to happen for certain classes of problems, including those in power systems known as “state estimation”.

In trying to develop a body of theory for nonconvex optimization in power systems, we ended up proving a general theorem with wide-reaching impliciations for artificial intelligence and machine learning. Professor Lavaei suggested sending these results to the prestigious Neural Information Processing Systems (NeurIPS) conference. Not only was our paper accepted, but it was also offered one of only 200 spotlights, out of 5000 total submitted papers.

The NeurIPS spotlight paper added a whole new dimension to my research profile. It certainly helped me in landing a faculty position at the University of Illinois, Urbana-Champaign.

Congratulations on accepting a faculty position at the University of Illinois, Urbana-Champaign! What excites you most about being a professor?

Thank you! I’ve wanted to become some sort of mathematician, scientist, or engineer since as far back as I can remember. Having achieved my own dreams, what excites me now is helping others to achieve their own. As a professor, I am most excited about the ability to help the next generation of mathematicians, scientists, and engineers discover their passions and succeed in their careers.

As a new professor, how do you think the OR field will change over the coming decades? What kinds of problems do you think IEOR research can help solve over this time period?

Our field is slowly but surely moving towards big data and large-scale computation. In this new world, IEOR research will one day offer a theoretical explanation behind this empirical success. IEOR researchers will one day give an answer to the question “why does machine learning work at all, and why does it work so well?”

What will you miss most about Berkeley?

I will miss the friends I’ve made at Berkeley, and all those times we worked into the late evening together while munching on reheated pizzas. I will also miss the breathtaking views of San Francisco and the Golden Gate Bridge from that parking lot behind the Lawrence Hall of Science.

Keith McAleer