Title: A Kernel Hike in the Machine Learning Forest

Abstract:

Machine learning is a dense forest with many trails and vantage points---the vantage points are the problems of interest, e.g., classification, regression, testing, sampling, dimensionality reduction, etc., and the trails are the algorithms that lead us to the vantage points by solving the problems of interest. I will take you on a hike (talk) through this deep forest's "Kernel Trail", which in contrast to many other trails there, offers several viewpoints. The trek is intended to showcase several viewpoints on this trail and is designed to be easy enough for beginners who are not experts. More breathtaking views (deeper understanding of problems) are available by further continuing on difficult and rugged portions of this trail but this would require a lot of training and preparation. I'll mention these challenges in passing and paint a picture of how breathtaking the vistas are.