Johanna Devaney, Conservatory of Music The project is motivated by the idea that humans are able to learn with greater efficiency than machine learning models, in large part because they learn not just from exposure, but also from domain knowledge, which includes guided practice and codified knowledge. For example, in music, people learn by both playing their instruments and classroom study of music theory. This work will develop new approaches for integrating domain knowledge into deep learning models. It will create models that can be trained with less data as well as mitigate data biases.
