Sirui Yao, DAC Ph.D. student in the Department of Computer Science

Graphic is from Yao’s NeurIPS 2017 paper “Beyond Parity: Fairness Objectives for Collaborative Filtering”

Sirui Yao studies the biases of recommender systems.

“A recommender will often suggest different courses to male and female college students because based on historical data, there are differences in course preference between these two groups,” said Yao, a Ph.D. student in computer science at the Discovery Analytics Center.

“Over-leveraging this gender-based pattern encourages stereotypes and creates an even bigger —and undesirable — gap between demographic groups, especially in areas actively encouraging equality, such as engineering,” she said.

Yao’s work proposes methods for measuring, analyzing, and mitigating unfairness in recommender systems. She was awarded a 2018-2019 Deloitte Foundation Data Analytics Fellowship in the amount of $10,000 to fund her research.

Her advisor, Bert Huang, introduced Yao to this topic three years ago. “I realized this is a very critical yet fairly unexplored area in machine learning, so I wanted to focus on it and make whatever contributions I can to fill this vacuum,” she said.

“Being a DAC student means I am always informed about exciting data science projects and surrounded by people who have a lot of expertise, passion, and creativity in data analytics,” Yao said. “Such an environment encourages me to learn more and do more.”

Yao works in Huang’s Machine Learning Laboratory and the two have collaborated on research, including “On the Need for Fairness in Financial Recommendation Engines,” which Yao shared at the NeurIPS 2018 Workshop on Challenges and Opportunities for AI in Financial Services: the Impact of Fairness, Explainability, Accuracy, and Privacy.

She presented “Beyond Parity: Fairness Objectives for Collaborative Filtering” at the main NeurIPS 2017 conference and  “New Fairness Metrics for Recommendation that Embrace Differences” at the KDD 2017 Workshop on Fairness, Accountability, and Transparency.

This past summer, Yao interned at Google Brain in New York City, where she worked on a research project that designs a trajectory simulation and analysis framework for studying the long-term dynamics of recommender systems. This research has been submitted to WWW 2020: The Web Conference.

Yao earned a bachelor of science degree in computer science and technology from the Harbin Institute of Technology in China and is projected to graduate from Virginia Tech in December 2020.