Course Identification

There are no formal prerequisites. However, the course requires mathematical maturity, and students are expected to be familiar with linear algebra and probability, as taught in undergraduate computer science or math programs.

This course will provide a self-contained introduction to some of the actively-researched areas in machine learning today. It will cover theoretical principles and challenges as well as practical algorithms. The focus will be on supervised and discriminative learning, where the goal is to learn good predictors from data while making few or no probabilistic assumptions. Along the way, we will introduce and use tools from probability, game theory, convex analysis and optimization. The course will cover the following topics (time permitting):

• Statistical learning: Statistical learning models; Overfitting; Generalization and sample complexity; Uniform Convergence; Stability; Linear predictors; Support Vector Machines
• Online learning and optimization: No-regret learning; Online convex optimization and gradient descent; Online-to-batch methods; Learning from Experts; Follow-the-Leader Algorithms
• Advanced topics -- this will include a subset of the following: Learning with information constraints and partial information; Deep learning; Learning on distributed systems; Advanced optimization algorithms for learning problems; Trade-offs between sample complexity and computational complexity

Upon successful completion of this course students should be able to:

1. Describe basic concepts, principles and algorithms in the field of machine learning.
2. Apply their acquired knowledge of machine learning methods and principles in their own areas of research.