Meet ML@GT: Zhanzhan Zhao Uses Machine Learning to End Residential Segregation in Atlanta

The Machine Learning Center at Georgia Tech (ML@GT) is home to many talented students from across campus, representing all six of Georgia Tech’s colleges and the Georgia Tech Research Institute (GTRI).

These students have diverse backgrounds and a wide variety of interests both inside and outside of the classroom. Today, we’d like you to meet Zhanzhan Zhao, a third-year machine learning Ph.D. student who is using machine learning to help end residential segregation in Atlanta.

Name: Zhanzhan Zhao

Advisor: Dana Randall

Major: Third-year Machine Learning Ph.D. student in Computer Science

Previous degrees earned and from what institutions: Bachelor’s degree from Harbin Institute of Technology and a master’s degree from University of Michigan, Ann Arbor

Tell us about your research interests. Where might people be impacted by them in everyday life?

I am interested in studying phase changes theoretically in segregation models. A particular focus has been on how infrastructure and socio-economic considerations can improve or worsen segregation.

We also use machine learning tools to answer questions about where and what kind of urban infrastructure placement, such as expansion topologies of transportation systems, would be ideal to mitigate residential segregation in Atlanta.

We hope that our work will be of use to the Atlanta-Region Transit Link Authority (ATL), which covers 13 metro counties and is working to improve Atlanta transit over the next 20 years.

What drew you to wanting to research these areas?

I think that the phenomenon of social division within a community, or segregation, is pervasive and problematic. Segregation helps form polarized minisocieties that propagate unfairness, create competition for resources, and promote uncooperative efforts to seek political authority.

As an individual, I often feel that it’s a pity that we hold invisible pride and prejudice over others that we rarely interact with on a personal level. Considerable social science literature also suggests that integrated heterogenous neighborhoods are more rewarding and robust than homogeneous ones. Because of all of this, I want to explore how to mitigate segregation by urban infrastructure and socio-economic considerations using statistics and machine learning tools.

What motivated or inspired you to pursue your Ph.D.?

My mom is a university professor, and I like that she has the freedom to allocate her time while also pursuing her intellectual interests. I also gained confidence in my research abilities from my experiences working with my master’s advisor Dennis Bernstein and other mentors from my time at the University of Michigan.

Earning your Ph.D. requires a lot of work and sacrifice. What has been challenging, rewarding or unexpected about this experience?

I think the transition from a master student to a Ph.D. student has been challenging for me. The Ph.D. phase is still kind of a weird mixed status of life. You’re in school as a student, mentee, and employee. Your family and society expect you to take on more responsibilities, but walking around campus makes me feel like I’m a kid in school all of the time. It has been confusing for me to identify and balance various expectations compared with life when I was a master’s student.

Why did you choose Georgia Tech for your Ph.D.?

I was attracted by the school culture, which is so lively, interactive, and open-minded. The new ideas and opportunities are abundant. I feel very lucky to get into the ML program and work with my advisor, Dana Randall. 

What’s your favorite Georgia Tech memory so far?

I miss the days in Coda where my good friends and I worked and had fun together. My best friend and I usually went to gym early in the morning, and ate breakfast together in the common kitchen to start the work day. We were also so happy playing the werewolf game together during the weekly ML social hour.

What have you been up to during quarantine?

I began to cook every meal myself at home and I have become a slightly better chef now. I also made some friends from social media. Through our talks, they have brought fresh perspectives and various life stories to me.

Though 2020 seems to have lasted a lifetime, we are still at the beginning of a new decade. What are you most looking forward to in the next 10 years?

I have no idea what the future will be like. For now, I am looking forward to writing a very good paper with my advisor!

What is your proudest accomplishment?

I guess it is that I published a research paper in a top journal (Journal of Guidance, Control, and Dynamics) in my former field of aeronautics. It has three citations so far!

What are some of your hobbies?

I like playing Chinese Chess and werewolf games with friends. Although I do not often win, I enjoy the feeling of competition, and I like surprising people by playing with unconventional strategies! 

Podcast, book, movie, or tv show and why? Of your chosen medium, what are some of your favorites?

Book, I guess. I like books because they can more easily convey abstract concepts and I can finish them quickly. I like a book called SuperCooperators: Altruism, Evolution, and Why We Need Each Other to Succeed. The author wrote about his research and it’s as exciting as a novel.

If you could time travel to any period of time, where would you go and why?

I would travel to the time when I was 18. I would like to tell the past me to be less ambitious about the future and to grasp happiness. I’d also tell her to be more sincere and devote more time to her friends.

What’s your favorite place to study or work and why?

Now it is home because it is quiet and safe. I can also Skype my best friend Neda so we can work together as well.

Why do you think embodying Georgia Tech’s motto of “progress and service” is important, especially in regards to ML and AI?

When we keep the spirit of service and progress in mind as students and researchers, it helps our work become more meaningful. It’s a good motivation and can be a good guide for our research.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.