Zhizhuo Zhou selected for NSF Graduate Research Fellowship

Computer science undergraduate Zhizhuo Zhou has been awarded a prestigious NSF Graduate Research Fellowship to support his upcoming graduate studies in Fall 2021.

The program recognizes and supports outstanding graduate students in NSF-supported science, technology, engineering, and mathematics disciplines who are pursuing research-based masters and doctoral degrees at accredited United States institutions.

Zhizhuo is interested in computer vision and its ability to make breakthroughs in interdisciplinary fields such as ecology and climatology. 

For his graduate research, Zhizhuo hopes to continue his research at the intersection of computer vision and natural language processing. Specifically, he wants to create intelligent systems that can be deployed in new domains with very little training data. His goal is to achieve human-level learning performance in vision and language tasks. Furthermore, he wants to help make machine learning accessible to everyone by sharing his work through video.

Zhizhuo is grateful for the mentorship and support he received from faculty in the department during his undergraduate studies.

In 2019, Zhizhuo worked on an air quality estimation project with Prof. Robert Dick which used large amounts of images on social media to predict air quality index at higher spatial resolution than physical sensor stations miles apart from each other. Because of a lack of real-world data, Zhizhuo was able to validate the algorithm using synthetic models of cities that he created.

Also in 2019, Zhizhuo co-founded the Michigan Alexa Prize Team which competed to build an open-domain conversational bot. The team was selected as a top 10 finalist out of over 380 university teams worldwide to receive a $250,000 research grant. Zhizhuo is grateful for the support he received from Prof. David Jurgens, who advised him for the competition.

In the summer of 2020, Zhizhuo worked with Prof. David Fouhey to quantify bird skeletons and study the effect of climate change on the evolution of bird species. For this project, he created a convolutional network of 16 layers in an encoder decoder structure that made it possible to use machine vision to accurately measure a large library of bird skeletons.

Additionally, Zhizhuo, in a team with five other undergraduates, worked on generating high resolution climate simulations using deep learning models in the ProjectX2020 machine learning for climate change competition. With mentorship from Dr. Sindhu Kutty, the team won the best paper award in the extreme weather category. 

Zhizhuo has served as a project lead on the Michigan Data Science Team, where he gave short, weekly presentations on data wrangling and data analysis for students without computer science backgrounds.