7/20/2021 Laura Schmitt, Illinois CS
Felipe Arias, Jaron Mink, Christopher Perdriau, and Seth Poulsen earned prestigious NSF Graduate Research Fellowships and will conduct research to address robot navigation, user online security, broadening participation in CS, and mathematical proof instruction.
Written by Laura Schmitt, Illinois CS
CS graduate students Felipe Arias, Jaron Mink, Christopher Perdriau, and Seth Poulsen have each received a 2021 National Science Foundation Graduate Research Fellowship, a highly competitive grant that provides a $34,000 annual stipend and $12,000 cost-of-education allowance for three years.
Interested in applying for a 2022 NSF Graduate Research Fellowship?
Start working towards the October 19, 2021 application deadline by reviewing the FAQs and the Program Solicitation at the NSF website.
Illinois CS offers resources to help students prepare their applications. Contact Hannah Gorrie, coordinator of constituent awards, for more information or visit the CS Graduate Fellowship wiki page.
The Graduate College's Fellowships Office has workshops and other resources to help students with fellowship submissions.
“An NSF Graduate Research Fellowship is a tremendous recognition for the recipient,” said CS department head Nancy Amato, an Abel Bliss Professor of Engineering. “In addition, the financial award provides the student freedom to pursue research topics of his or her interest. I encourage all eligible CS students—seniors and first- or second-year graduate students who are U.S. citizens or permanent residents—to apply for one of these awards. Our department and the Graduate College have excellent resources to assist students in developing competitive applications.”
A third-year doctoral student in professor Nancy Amato’s research group, Arias is improving mobile robot navigation by developing algorithms that consider a robot’s surroundings, experience, and relation to other agents when moving in constrained pedestrian environments such as hospitals, schools, or homes.
Arias recently published a paper at the premier international robotics conference (IEEE ICRA 2021) outlining a motion planning method that enables groups of robots to efficiently compute paths and avoid collisions. His approach includes a sampling-based multi-agent motion planning method and a self-supervision methodology for training data generation, which he developed in collaboration with researchers from Google Brain.
Before starting the doctoral program at Illinois, Arias gained valuable research experience as an undergraduate researcher at Berkeley and Stanford, where he developed a phone-based incident reporting system for a mobile app and worked on weak supervision for natural language processing, respectively. In addition, Arias worked on applications for the research project that is now Snorkel AI, a startup that handles the data and labeling needs of companies working on difficult machine learning problems.
He also worked with Illinois Industrial and Enterprise Systems Engineering professor Richard Sowers’ research group on applying machine learning techniques to track human behavior in the roadways, assess risk, and detect areas that may require more law enforcement or traffic signals.
Arias has received several other awards for his research and academic performance, including a 2020 National GEM Consortium Graduate Engineering and Science Fellowship, a 2019 Illinois CS Saburo Muroga Endowed Fellowship, the 2019 Grainger College of Engineering C.S. Larson Transfer Student Award and Scholarship, and an Engineering Pathways Scholarship (2016-19).
After earning his PhD, Arias plans to pursue a career as a research scientist and entrepreneur. Originally from Bolivia, Arias immigrated to the United States when he was 12 and is passionate about supporting the education and needs of fellow immigrants and Hispanics with an interest in STEM careers.
“This fellowship is my biggest accomplishment so far and something I have been thinking about for over three years,” said Arias. “[Winning] has been a bit surreal, but also just another reminder that hard work pays off. Now that I have more academic liberty, I hope to pursue topics outside of the scope of my previous funding sources and spend more time on a non-profit organization I am putting together.”
Jaron Mink, who is a member of CS assistant professor Gang Wang’s research group, is investigating how people behave and judge risk during privacy-sensitive actions online. Tools that are designed to protect the security and privacy of users and companies may prove less effective because researchers hold unrealistic or misguided expectations of real user behavior.
Specifically, Mink is examining whether people trust the classification decisions of machine learning models in high-risk security situations—for example, determining whether a file contains malware.
Although many machine learning explanation techniques have been proposed, few have evaluated whether users can understand and make use of these explanations, said Mink, who is starting his third year of graduate school this fall. His work will see whether current machine learning explanations are appropriate for these high-risk situations and what possible improvements to explanations may be required.
Mink possesses a unique blend of industry and academic research experience. As an undergraduate at UCLA, he worked as a software consultant for an IT security company, where he helped develop a mobile application for online autism treatment.
He also worked for a broadband company, where he and his fellow interns created a dashboard to aggregate, analyze, and highlight notable security alerts. This dashboard made it easier for network administrators to know which alerts required more in-depth investigation.
During his first year of graduate studies at Illinois, Mink worked with assistant professor Adam Bates on the development of an attack-preserving log reduction method that retains all attack-relevant logs. He presented this work at the Annual Computer Security Applications 2020 conference. He is also actively conducting a study on fitness app user behavior that investigated the usage and perception of location privacy mechanisms on shared fitness routes.
This summer, Mink is performing research at the Max Planck Institute for Privacy and Security. Ultimately, he aims to pursue a career in academia as a professor.
“This award reminds me that what's seemingly insurmountable is possible with patience, focus, and the support of good people in your life,” Mink said. “Since my domain of human factors is inherently interdisciplinary, the flexibility this funding provides will allow me to better extend my network and work with collaborators [that] I might not have been able to otherwise.”
Christopher Perdriau, who earned his BS degree in CS from Oregon State University in June, is identifying and understanding the barriers and stereotypes that prevent certain groups of people from participating in computer science. Working with CS assistant professor Colleen Lewis, he aims to make the CS field more equitable and accessible to a broader range of people, while creating more inclusive software and career opportunities for under-represented groups.
Perdriau was drawn to the human-computer interaction and inclusive design areas of research because of personal experiences growing up. His parents, who are both deaf, were excluded from fully participating and enjoying his school activities because of technology limitations. In high school, he also noticed the lack of women students in his CS courses; he himself often felt like an outsider in a programming class dominated by fellow students with extensive coding experience.
At Oregon State, Perdriau met an innovative faculty member who introduced him to GenderMag, which is used by software teams around the world to identify gender-inclusiveness issues in software interfaces. He conducted multiple research studies on GenderMag and InclusiveMag, which engineers use to design products that are gender inclusive.
He continued conducting research this summer at the University of Washington, studying the recruiting practices that informal CS programs like summer camps, after-school clubs, and hackathons use to recruit students and how those practices impact who get the opportunities to participate in these CS programs.
“Winning this fellowship means that I’ll be more financially stable, allowing me to spend more of my energy, time, and focus on research,” said Perdriau. “I’ll be held to the same standards as past fellows, which is inspiring and an honor because of the research they have conducted.”
During his doctoral studies, Perdriau aims to secure an industry internship so he can determine whether he wants to pursue an academic or industry career after earning his degree.
Entering his third year of the doctoral program, Seth Poulsen is exploring ways to improve how instructors teach students to write mathematical proofs. ACM identified proofs as a core knowledge area for students majoring in CS, computer engineering, or software engineering.
Poulsen developed Proof Blocks, which allows students to construct mathematical proofs by dragging and dropping prewritten proof lines into the correct order, rather than having to write the entire proof from scratch. Poulsen evaluated Proof Blocks problems by using them for exams in a U of I discrete mathematics course.
He describes Proof Blocks in a paper that he co-authored with his faculty co-advisors Geoffrey Herman (CS) and Matthew West (MechSe) and CS professor Mahesh Viswanathan. Poulsen will present the paper at the 2021 International Computing Education Research conference, where it was nominated for a best paper award.
Poulsen earned his bachelor’s degree in mathematics from BYU in 2018, graduating summa cum laude. He then worked as a software development engineer at Amazon, where he contributed to the launch of the Kindle Lite mobile application that allows millions of people to read E-books on their phones.
As an incoming doctoral student at Illinois in 2019, he received an Andrew & Shana Laursen Fellowship.
“I’m very grateful for the fellowship because it means I can focus on my high-impact research for the next few years without needing to worry about funding,” said Poulsen.
After graduating, he hopes to pursue a tenure-track faculty position so he can continue his work teaching computer science and doing education research.
In all, the U of I campus had 27 students earn 2021 NSF Graduate Research Fellowships. Three other CS students earned honorable mention honors for the Fellowships— Ruta Shirish Jawale, Nerla Jean-Louis, and Kevin Johannes Ros.
About the NSF GRF program
Launched in 1952, shortly after Congress established NSF, the Graduate Research Fellowship (GRF) program represents the nation's oldest continuous investment in the U.S. scientific workforce. The program recruits high-potential, early-career researchers and supports their graduate training in science, technology, social science, engineering, and mathematics fields. In addition to receiving an annual stipend and cost-of-education allowance, awardees have access to two professional development programs: the Graduate Research Opportunities Worldwide (GROW) program, which supports students researching in overseas labs, and the INTERN program, which supports research internships in any sector of the U.S. economy. Fellows also have access to NSF’s Career-Life Balance initiative. For additional details, visit the NSF-GRF webpage.