Two UMD Undergraduates Receive Honorable Mentions in CRA Outstanding Undergraduate Researcher Program

Two University of Maryland undergraduates have been recognized with honorable mentions from the 2021 Computing Research Association (CRA) Outstanding Undergraduate Researcher Award program.

The prestigious program—sponsored this year by Microsoft Research—recognizes undergraduate students in North American colleges and universities who show outstanding potential in an area of computing research.

Neehar Peri (left in photo), a senior majoring in computer engineering, and Ian J. Costello (right), a junior who is double majoring in computer science and mathematics, are among the 105 students being recognized by the CRA this year. There were four winners, five runner ups, 23 finalists and 73 honorable mentions.

CRA said in a statement on their website that this year’s nominees are a “very impressive group,” with many of the students involved in more than one research project, and others active in the development of software and apps.

CRA is an association of more than 200 North American academic departments of computer science, computer engineering, and related fields; laboratories and centers in industry, government, and academia engaging in basic computing research; and affiliated professional societies.

Peri is conducting research with John Dickerson, an assistant professor of computer science, and Rama Chellappa, a College Park Professor of electrical and computer engineering. Dickerson and Chellappa both hold appointments in the University of Maryland Institute for Advanced Computer Studies (UMIACS).

Peri’s work with Dickerson focuses on building more robust machine learning systems. Specifically, he has designed defenses for clean-label poisoning attacks, a type of adversarial attack on the training data used for creating machine learning algorithms. More recently, Peri has begun working on building robust but fair auction mechanisms using techniques from deep learning.

In Chellappa’s research group, Peri is focused on building intelligent traffic analytics systems for next-generation smart cities. Specifically, he is working on improving privacy-aware vehicle re-identification and traffic anomaly detection.

Costello is conducting research with Abhinav Bhatele, an assistant professor of computer science with an appointment in UMIACS, in the Parallel Software and Systems Group, which Bhatele leads.

Costello says he is working on two primary projects with Bhatele. The first uses machine learning to predict the performance of scientific applications on supercomputers to optimize resource allocation.

Additionally, the group is building a massively parallel epidemiology simulation that will be able to model real behavior (via cell phone or other data sources) on a country-level scale to test the efficacy of various COVID-19 intervention strategies.

—Story by Melissa Brachfeld