10: Nonlinear Feature Transform (58min)

Machine Learning From Data, Rensselaer Fall 2020. Professor Malik Magdon-Ismail talks about the nonlinear feature transform. We focus on the mechanics of the nonlinear transform which allows us to use the power, efficiency and convenience of linear models while benefiting from nonlinear classification boundaries. Of note is that with this immense power comes immense responsibility to use it carefully so that you maintain the link between in-sample and out-of-sample error. This is the tenth lecture in a "theory" course focusing on the foundations of learning, as well as some of the more advanced techniques like support vector machines and neural (deep) networks that are used in practice. Level of the course: Advanced undergraduate, beginning graduate. Knowledge of probability, linear algebra, and calculus is helpful. Material is from Chapter 3 of "Learning From Data", amlbook.com, 2012.