The project aims to develop an automated machine learning (AutoML) pipeline to accelerate the development of surrogate models for predicting materials properties. Working with tabular data linking composition of an alloy to material properties, the intern will design tools for data curation, composition-based featurization, feature space reduction, and output transformation. A core outcome will be a modular library of ML models and an AutoML system that selects optimal pipelines based on data characteristics. A final pipeline will include the implementation of the AutoML workflows in a probabilistic framework suitable for integration with a Bayesian optimization engine, supporting faster and more accurate materials design.
This record is a historical disbursement from Canada's open grants database — it documents a past payment, not an active program accepting applications.
View disbursement record →