Introduction to deep learning: A first course in machine learning

Abstract

A new undergraduate course on deep learning is described. Most of the students who took the course were junior and senior computer science majors. Nearly all the students in the course had not had a previous course in machine learning. The course builds on basic concepts students learn in calculus, statistics and probability courses. Key concepts from machine learning, e.g. the cardinal sin of over-fitting, are introduced in the context of deep learning, in a problem driven manner, so that students discover and observe these concepts for themselves. A concept map as well as useful online resources are described in the appendix.