Xiao Wang Receives Google Research Scholar Program Award

Due to the sensitive nature of user data regularly harvested by companies, regulations like the California Consumer Privacy Act and the General Data Protection Regulation restrict how this data is stored, used to train algorithms and machine learning (ML) models, and applied in decision-making processes. This poses a problem for auditors charged with ensuring data collection and usage compliance.

Xiao WangNorthwestern Engineering’s Xiao Wang has received a 2023 Google Research Scholar Program Award for his work developing zero-knowledge proofs (ZKP), which allow a party to verify a statement without revealing the underlying information that “proves” the data. The ZKP protocols are critical cryptographic tools to enable transparent and policy-compliant use of confidential data.

Wang is an assistant professor of computer science at the McCormick School of Engineering.

The Google Research Scholar Program supports the advancement of world-class research by early-career faculty members in fields including algorithms and optimization, human-computer interaction, ML and data mining, natural language processing, privacy, quantum computing, security, and systems. Through the program, Google aims to facilitate connections among junior faculty and encourage the formation of long-term collaborative relationships.

Wang will receive $60,000 for use during the 2023 academic year for his project titled "Zero-Knowledge Proofs for Private and Transparent Machine Learning." The work will focus on using ZKP protocols to improve the transparency of the ML algorithm without revealing information about the underlying data used in the algorithms.

Wang aims to develop systems that will allow companies to prove how its machine-learning models are trained and used without showing any details of the user data. His preliminary research demonstrates the practical application of this approach in selected models, and the goal of this project is to extend its efficiency and applicability to other models and properties.

“The protocols designed from this award will improve the efficiency of zero-knowledge protocols used for machine learning and can potentially address the issues in ensuring ML algorithms are properly used in practice,” Wang said.

Wang explained that the high computational and communication costs of ZKP currently prevents its use in large-scale applications. He aims to improve the foundation of scalable ZKP to use them to enhance the security and transparency of modern machine-learning tasks.

“I am very honored to receive a Google Scholar award and would like to thank the Department of Computer Science for its continuous support,” Wang said.

Xang recently received a National Science Foundation Faculty Early Career Development Program (CAREER) award to support his research advancing the scientific foundation of secure multi-party computation.

Prior to joining Northwestern, Wang completed a post-doctoral fellowship in computer science at MIT and Boston University in 2019. He earned a PhD in computer science from the University of Maryland, College Park, and a bachelor of engineering in computer science from Hong Kong University of Science and Technology.

McCormick News Article