Mathematical Sciences Department Statistics Seminar
Jake Gagnon, Biogen Inc.
Thursday, April 10th
2:00pm - 3:00 pm
Title: ML Best Practices with Reproducible Reporting
Abstract: In this talk, we will discuss best practices in traditional machine learning with illustrative examples in pharma, medicine, and omics. The best practices begin with pre-processing of the data, assessment of the data, and selection of a few candidate models. Afterwards, we will illustrate model evaluation and comparison using data splitting techniques and model diagnostics. Finally, we conclude the talk by advocating for reproducible research with software such as R Quarto.