PCS Framework, Interpretable Machine Learning, and Deep Neural Networks

In this talk, Dr. Yu will discuss the intertwining importance and connections of three principles of data science: predictability (PCS), computability and stability (workflow and documentation). She will also define interpretable machine learning through the PDR desiderata (Predictive accuracy, Descriptive accuracy and Relevancy) and discuss stability as a minimum requirement for interpretability. PCS and PDR will be demonstrated in the context of one collaborative project in neuroscience, DeepTune, for interpretable data results and testable hypothesis generation.