Project Development of thermoplastic starch aerogels using an adaptive experimental design in a Bayesian optimization framework Program Engineering Term Academic Year 2024 Status Accepted Research Area Materials, Machine Learning Description This project focuses on synthesizing and optimizing thermoplastic starch (TPS) aerogels. These compostable, ultra-lightweight materials are used in several applications, including thermal insulation in rockets, balloons, and other aerospace vehicles. The specific objective of this project is to develop optimal TPS aerogel formulations and processing protocols leveraging an adaptive experimental design approach that uses variational Gaussian process regression in a Bayesian optimization framework. Our group also seeks applications in Motorsports Engineering. Supervisor Andres Tovar Mentor Type of work required The project involves work in three areas: material synthesis, Bayesian optimization, and material characterization. Material synthesis and characterization will be performed in our materials research laboratory. Bayesian optimization methods incorporating variational Gaussian process regression and other machine learning approaches will be developed in Python. Websites https://www.compostablept.com/ Qualifications Students involved in this project should have excellent analytical skills, attention to detail, and clarity. Taking courses or having previous knowledge in chemistry, materials, and programming is desirable but not required. Credit Hours 2 Weekly Hours 5 (estimated)