Q&A

What is the research you are engaged in and what have you accomplished this summer?

The goal of my project is to improve the precision of the electron beam position reconstruction at the Cornell Electron Storage Ring (CESR) by creating a predictive model that utilizes neural network artificial intelligence. This summer, I have built a pipeline to produce a non-linear regression model that minimizes error, involving data processing and engineering, neural network building, network training and optimization, and performance evaluation. So far, I have optimized a neural network that predicts the beam position to an accuracy within 1 micron using idealized simulated data. The next step for this project is to now inundate the data with sources of error inherent to the beam position monitoring system, and from there, to improve the turn-by-turn precision of the beam position that the neural network model reconstructs.

What are the impacts of your research for the general public?

The sources of error inherent to the machinery of the CESR beam position monitoring system set the limit for how well the position of the beam can be determined as it passes through the electron ring. The beam position precision is a core factor that limits the performance of CESR and CHESS (Cornell High-Energy Synchrotron Source), and thus limits the performance of the research that can be done at these facilities. The X-ray research done at CHESS, for example, can be applied to fields from biology and medicine to materials science, and the quality of those results depends on the quality of the X-ray beam produced at CESR.

Additionally, this project serves as another example of how neural networks can be used to create models that can make predictions based on input data, which can be applied to many fields of research, engineering, and data science. On a broader scale, it is important to understand the capabilities and limitations of neural network AI to create predictive models that are effective and ethical.

What did you enjoy most about this research/summer experience?

I have really enjoyed learning about neural networks because that has always been a 'bucket list' goal I have wanted to do ever since first hearing about them. It was also exciting to learn about accelerator physics since I have never been involved in this side of research before. I enjoyed seeing my skill level improve in something that was very new to me and producing results that I had set a goal for at the beginning of the summer.

What did you find the most challenging about this research?

The biggest challenge for me was being able to present my work and effectively communicate the goals for my project and my results. During the REU, I've had to present my research fairly frequently to teams of experienced researchers as well as other REU students. Sometimes I get nervous about presentations, but it becomes easier with practice.

How has this experience changed your view about being a researcher?

This REU program has highlighted for me the importance of interdisciplinary research. I was worried I wouldn't be accepted to this REU program because my studies and previous research projects have been focused on astrophysics, but this experience has taught me that my contributions are also welcomed and that diverse skill sets between researchers can bring new insights to a project.