Kaiyang Zhao, a Ph.D. student in Carnegie Mellon University's Computer Science Department (CSD), was selected for the North American Qualcomm Innovation Fellowship.

Zhao's project, "Learned Virtual Memory for Heterogeneous Architectures," aims to radically rethink virtual memory using lightweight machine learning models to solve challenges in data centers and at the edge.

Virtual memory has become a severe bottleneck due to the significant increase in memory-intensive workloads such as artificial intelligence applications. The ambitious goal of the project is to provide a pliable abstraction that dynamically learns and adjusts based on the workload behavior and underlying hardware. The result will enable major performance and energy efficiency gains, and ultimately reduce the carbon footprint across edge devices and servers in data centers.

Zhao will collaborate with Yuang Chen, a Ph.D. student in the Electrical and Computer Engineering Department in the College of Engineering. Both students are part of CMU's Computer Architecture and Operating Systems (CAOS) group and are advised by Dimitrios Skarlatos, an assistant professor in CSD with a courtesy appointment in ECE. 

Zhao and Chen make up one of two CMU teams selected for the fellowship. Ruben Purdy and Wei Li, both Ph.D. students in ECE, were selected for their project, "Improving Diagnosis With Physically Aware Fault Models."

Students selected for the Qualcomm Innovation Fellowship earn a one-year fellowship and are mentored by Qualcomm engineers to facilitate the success of the proposed research. The fellowship comes with $100,000 to fund that research.

The Qualcomm Innovation Fellowship has awarded more than $20 million in funding since 2009. The fellowship receives nearly 400 submissions each year and funds about 45 students annually. This year there were 16 winning teams in North America.

Learn more about the Qualcomm Innovation Fellowship on the company's website.