Student Spotlight: Harry Dong
Monica Cooney
Jan 11, 2024
Learn more about research team member, Harry Dong, who recently received the Wei Shen and Xuehong Zhang Presidential Fellowship.
Educational Background:
Ph.D. Candidate, Electrical and Computer Engineering
Statistics & Computer Science BA, UC Berkeley, 2021
Statistics & Computer Science BA, UC Berkeley, 2021
Interests outside the classroom/lab:
Racquetball, Dance, Tennis, Reading
Research thrust in Center:
What does your research entail?
My research in the Center involves adapting state-of-the-art deep learning methods to materials applications, particularly in microscopy. I seek to leverage the power of generative models such as large language models and diffusion models to better understand, enhance, and generate materials data while also being mindful of the underlying physical properties, constraints, and materials data scarcity. The impacts of such include accelerating microscopy data collection and recovering missing or corrupted data.
Within the Center, I work closely with Megna Shah, Sean Donegan, and Jeff Simmons. Their domain expertise guides more principled approaches to model design to be tailor made for such problems. More broadly, I have an interest in efficient deep learning. As models become more computationally demanding, efficiency becomes a more pressing issue. By exploiting underlying model or data dynamics, I hope to make deep learning more accessible.
How has being part of the center impacted your research?
Being part of the Center has allowed me to see unique problems and challenges in the growing intersection of materials science and machine learning, where each field can bring new ideas to the other.