Introduction to Machine Learning

Graduate / Professional

This course focuses on fundamental topics in machine learning. We begin with an overview of different components of machine learning and types of learning paradigms. We introduce a linear function, discuss how one can train a linear function on a given dataset, and utilize it to tackle classification and regression problems. We then consider kernel methods to enable us to solve nonlinear problems. Additionally, we introduce the concept of generalization error and overfitting. We discuss the role of regularization and extend linear regression to ridge regression. We also cover topics in optimization, beginning from gradient descent and extending it to stochastic gradient descent and its momentum variant. We also cover the concept of alternating optimization and topics within it. We introduce the curse of dimensionality and discuss topics on dimensionality reduction. Finally, we conclude the course with neural networks: how to build them using the topics discussed, how to optimize them, and how to apply them to solve a range of machine learning tasks. Prerequisites: Courses in data structures and object-oriented programming (e.g. CPSC 223a or equivalent courses), foundational mathematical tools such as discrete math and linear algebra (e.g. CPSC 202 or equivalent courses), calculus (e.g. MATH 112, MATH 115, MATH 120, or equivalent courses), linear algebra (e.g. MATH 225, or equivalent courses), and artificial intelligence (e.g. CPSC 370/570). A background in statistics is useful but not required. Experience in programming with Python and familiarity with Google Colab and numerical and image processing packages (i.e. NumPy, SciPy) is helpful. TTh 1:05pm-2:20pm