CHDS visitor and Williams College student David Krane spoke at the CHDS lunchtime seminar series on “Using Simulation to Improve Machine Learning of Complex Systems”.
Predicting the way in which a given cancer treatment will affect a tumor can be seen as a machine learning problem, Krane discussed. This prediction is a unique challenge because of complexities of biological systems, the number of potentially relevant genomic and clinical features, and the absence of large-scale patient data repositories. In the seminar, Krane presented a novel method to integrate simulations — a strong form of prior knowledge — into machine learning.
Learn more: Learn about decision science.
Related news: PhD Student at MGH
Related news: Modeling Software Seminar