COMP 562. Theory of Machine Learning.
Note: For information about Fall 2025 and Winter 2026 course offerings, please check back on May 8, 2025. Until then, the "Terms offered" field will appear blank for most courses while the class schedule is being finalized.
COMP 562
Theory of Machine Learning.
Credits: 4
Offered by: Computer Science (Faculty of Science)
This course is not offered this catalogue year.
Description
Concentration inequalities, PAC model, VC dimension, Rademacher complexity, convex optimization, gradient descent, boosting, kernels, support vector machines, regression and learning bounds. Further topics selected from: Gaussian processes, online learning, regret bounds, basic neural network theory.
- Prerequisites: COMP 330, COMP 350, COMP 360 and MATH 323
- Prerequisites: MATH 462 or COMP 451 or (COMP 551, MATH 222, MATH 223 and MATH 324) or ECSE 551.
- Restrictions: Not open to students who have taken or are taking MATH 562. Not open to students who have taken COMP 599 when the topic was "Statistical Learning Theory" or "Mathematical Topics for Machine Learning". Not open to students who have taken COMP 598 when the topic was "Mathematical Foundations of Machine Learning".
Most students use Visual Schedule Builder (VSB) to organize their schedules. VSB helps you plan class schedules, travel time, and more.
Launch Visual Schedule Builder