Presented by Nathan Crock, Department of Scientific Computing (DSC)
To build intuition about machine learning we need only understand optimization with one extra ingredient, generalization. The field of optimization rests upon three fundamental ingredients: a task or function, a way to measure performance at that task, and instructions on how to modify the function based on its performance. What makes machine learning different from optimization is the notion of generalization - the ability to perform well on data never before encountered. We will use real examples from different disciplines, with programs being executed in real-time, to understand optimization and then subsequently machine learning. Every example will be related back to the fundamental mathematical construct of machine learning - empirical risk minimization. By the end of the talk, attendees will have an intuition of the core concept behind machine learning, its capabilities, and most importantly, its limitations.
Tuesday, February 25 at 3:00pm to 4:00pm
Dirac Science Library, Dirac Instruction Room
110 North Woodward Ave, Tallahassee, FL, Tallahassee, FL https://www.lib.fsu.edu/dirac-science-library