Machine Learning and Compilers: A Survey of ML Techniques for Enhancing Optimizing Compilers / The Convenience of Pollution: The Struggle over Gasoline-Powered Leaf Blowers in the United States

Steele, Ryan, School of Engineering and Applied Science, University of Virginia
Norton, Peter, EN-Engineering and Society, University of Virginia
Morrison, Briana, EN-Comp Science Dept, University of Virginia

Energy use from unclean sources in the United States exacerbates global climate change. Strategies to reduce energy use can come from both technological innovation and behavioral changes. How might machine learning be exploited to improve compiler efficacy and efficiency? Research indicates that machine learning integration in compiler design may yield improvements to program efficiency, decrease compile times, and help automate the creation of optimization heuristics for different target architectures. Gasoline-powered leaf blowers (GLBs) emit large amounts of pollutants due to design limitations. How do advocates and critics of GLBs in the United States advance their agendas? Many homeowners oppose GLBs for their effects on health and the environment, while landscaping companies and manufacturers rely on them as business necessities. Participants in the debate have been divided by conflicting perceptions of personal responsibility to environmental stewardship and how best to balance regulation and personal freedoms.

BS (Bachelor of Science)
Energy use, Machine learning, Optimizing compilers, Gasoline leaf blowers

School of Engineering and Applied Science
Bachelor of Science in Computer Science
Technical Advisor: Briana Morrison
STS Advisor: Peter Norton

Issued Date: