---

The Interview:

Karen: What is your current research about?

Andrew: My research with Prof. Susu Xu focuses on enhancing the safety of autonomous vehicles, with a particular emphasis on pedestrian safety. A crucial task for autonomous vehicles is predicting the future trajectories of pedestrians, which is best achieved through deep learning models. However, the problem of bias has emerged as a challenge in the development of general AI systems, which tend to perform better on certain groups or demographics. In our research, we investigate whether the models that predict pedestrian trajectories are also subject to such biases. Previous studies have shown that pedestrians exhibit different walking patterns, which could affect the generalizability of these models. Our findings reveal that the state-of-the-art models we tested exhibit similar performance for men and women, but they perform worse for children and the elderly compared to the adult population. Addressing such biases is crucial for ensuring the safety and efficacy of autonomous vehicles in diverse settings.

Tell me how you got started in research at Stony Brook.

My first research experience was actually not with Prof. Xu, but with Prof. Benjamin S. Hsaio in the Department of Chemistry back when I was a sophomore majoring in Chemical Engineering. We were using advanced X-ray scattering techniques to discover more about plant-based materials. I was responsible for data analysis, which allowed me to develop my analytical skills. Additionally, I participated in a side project with Professor Yuefan Deng over the summer of 2021 after taking his class AMS 326: Numerical Analysis. I was fascinated by some of the concepts of numerical computing and wanted to learn more. Professor Deng invited me to assist in improving some of the instructional materials for the course, such as lecture notes, practice problems, and conceptual figures. This opportunity allowed me to enhance my understanding of the subject while also contributing to the education of future students. 

Although I was one of the top students in the Chemical Engineering department, I felt unfulfilled and disinterested in both my coursework and undergraduate research projects. It became clear to me that I wasn’t really that passionate about the field of chemical engineering. I realized that I needed to explore new areas of study, so I spent countless hours watching talks given by professors in various fields, hoping to find a topic that resonated with me.  Artificial intelligence (AI) was a topic that quickly drew my attention. My first instinct was to work on AI applications in chemical engineering, but I realized that combination was not exciting to me; I wanted a completely new start. Hot topics in AI like computer vision and natural language processing were interesting, but I found that recent papers in those fields focused heavily on theory; I was looking for something more applied. Eventually, I came across the field of smart transportation, which offered an opportunity to apply machine learning in a practical and meaningful way. I decided that I wanted to try it out for myself. After leaving Prof. Hsiao's group, I began working with Prof. Susu Xu in Fall 2021 to delve into this new field.

Was that difficult to switch gears, and start working in a brand new area?

Switching fields was undoubtedly one of the most challenging decisions I have ever made. Having invested so much time and effort into becoming a chemical engineer, the idea of starting from scratch in a completely new field was daunting. I was even told that I would never succeed in my new field ... However, looking back at all that I’ve accomplished so far, the gamble I took has really paid off. I’ve ended up finding a passion for researching smart mobility, and I think it’s a really exciting time to be in the field. There is a real opportunity to revolutionize our transportation systems. I’m really excited about what the future holds. 

It’s great that you’ve discovered what problems are of  most interest to you.

Yes, it’s been quite the journey. Although it was tough, I’m so fortunate to have met people along the way who have guided me in the right direction. I’m currently a senior, but I’ve decided to stay an extra year of undergrad in order to gain more research experience in the field before pursuing a PhD. I hope to continue researching ways of applying machine learning to various transportation and urban systems in graduate school.

Will you be focusing on PhD programs in computer science? or civil engineering?

I think that these traditional subject boundaries like “computer science” or “civil engineering” are becoming outdated in today’s world. The most pressing problems we face are often very complex and require interdisciplinary approaches, involving techniques and insights from a variety of different fields. As such, when applying to PhD programs, I will prioritize finding potential advisors whose research aligns with my interests, rather than focusing solely on the department name. There are many professors from civil engineering, industrial engineering, and computer science working on the kinds of real-world problems that excite me.

What are your plans for the summer?

Tell me about your experience of presenting off campus last semester. Did you enjoy the experience?

Yes, last fall I published a first-author workshop paper in SenSys 2022, a big annual conference on smart sensing and machine learning which was held in Boston last year. I also gave an oral presentation at the conference. Going through the process, I think the main takeaway was you have to learn how to give a big picture overview and package your work in a way that's presentable, organized, and effective. What have other people done previously? What is something new that you’ve contributed? Why is your new work important? Going through the process of organizing my work into a paper and preparing a presentation was a very valuable experience.

How beneficial was it for you to participate in the URECA Summer Research Program?

Participating in the URECA summer program was very useful. While you can set goals about how much research you will get done during a semester, these goals are never met from my experience.  Classes and other stuff that you have to deal with as an undergraduate student get in the way. But during the summer, you can focus on your research 110%, and that really accelerates your productivity and progress. 

What advice would you give about research to other students?

There are countless pieces of advice I could offer, but I want to really focus on one thing: embrace the uncertainty and try new things. Don't just overcome your fear of the unknown — actively pursue opportunities that pique your interest. Your undergraduate years are the perfect time to explore different fields and discover what you truly excel at and enjoy, as the stakes are lower now than they will be later in life. 

My own journey in research has been quite unique in that I switched fields quite late in my undergraduate career because I wasn't happy with the career path I was on previously. I know  that there are other students out there who are in a similar situation, but are hesitant to switch. To those people, I would say: there is never a better to time explore and pursue your passions than right now. At the end of the day you’re the one who knows yourself the best and you’re the one who’s in control of your future.