A Hitchhiker's Guide to Machine Learning
Host Campus: SUNY Polytechnic Utica
Dates: Thursday's, November 8, 15, 29, 5:30pm-7:30pm
Facilitator: Dr. William Thistleton, Associate Professor of Mathematics, SUNY Polytechnic Institute
Dr. William Thistleton is an Associate Professor of Mathematics at SUNY Polytechnic Institute. He served in the Peace Corps as a high school math and physics teacher in Cameroon, before earning his Ph.D. in Applied Mathematics at SUNY Stony Brook. His early work concentrated on computational fluid dynamics and ground water modeling, which led him to work in data analysis. Recently, he developed and presented a successful course in Time Series Analysis for Coursera. His teaching, research, and consulting interests include topics in data science such as image processing, machine learning, probability models, applied statistics, numerical differential equations, and mathematical modeling. He has taught over 20 different college courses in mathematics, and has a particular passion for teaching Calculus I to help ensure solid fundamentals in his students. He also regularly teaches graduate courses in Probability, Regression, Time Series Analysis, and Design of Experiments.
Professor Thistleton has long been involved with traditional data analysis approaches as a teacher, researcher, and consultant. Recently, like many data analysts, he has started engaging data sets and questions that need to go beyond these traditional techniques into machine learning. For instance, he is currently working on projects using machine learning to automatically identify chemical species in harsh, high temperature environments, and understand how sensors perform these cases. Last spring, he worked with several Master Teachers in the Mohawk Valley region to develop the structure for this mini-course.
This course will introduce you to the basics and some everyday applications of machine learning. Our lives are increasingly and profoundly impacted by data analytic techniques. If we want to talk about how cars can drive themselves or how Siri knows which flight you should book, we need to understand something about machine learning. These approaches are used in automating processes we use through better search algorithms and tagging images.
We will also discuss the similarities and differences between Artificial Intelligence, Big Data, Deep Learning, and Machine Learning. We will develop some of the big ideas and approaches in Machine Learning, such as classifiers (is that a picture of my Aunt or the Empire State Building?), Regression (how much should we ask when we list our house?), and Ranking (what should we watch tonight?). Basic data analytic techniques like Neural Networks, Decision trees, Random forests, genetic algorithms, and clustering will be explored. Our approach will be light-hearted and practical, and we’ll develop machine learning “hands on activities” for ourselves and for our students.
This mini-course is suitable for teachers with an undergraduate degree in science or mathematics and would like an overview of the topic. It will include highlights of Professor Thistleton’s research as well as general background. Although the topic may not be included in high school science courses per se (yet!), it should be of interest to both teachers and students as an example of computer science in our everyday lives.