AMATH 646Introduction to Mathematics of Deep Learning
- Units
- 0.50
- Level
- Graduate
- Faculty
- Mathematics
- Subject
- Applied Mathematics
When it runs
- F24
- W25
- S25
- F25
- W26
- S26
- F26
- W27
- S27
Ran in 0 of the last 7 terms. Winter 2027: scheduled, 1 lecture section. Spring 2027: schedule not out yet.
This course introduces the basic concepts of deep learning in a mathematically rigorous fashion. Topics include mathematical definitions of deep neural networks, approximation theory, gradient-based optimization, and generalization analysis. (Held with AMATH 446. Students in AMATH 646 will be expected to meet some additional learning objectives.)
Prerequisites
None listed.
Antirequisites
- Not completed nor concurrently enrolled in:
Leads to
No course lists it as a requisite.
Winter 2027 sections
| Section | Class | Enrolled | Meets |
|---|---|---|---|
| LEC 001 | 7527 | 0 / 25 | MW 14:30–15:50 |
Check it elsewhere for Fall 2026
- Graduate Calendar, Fall 2026The official entry
- Requirements for Fall 2026Waterloo’s requisite checker
- Schedule of Classes, Fall 2026Instructors, rooms, live seats
- Quest Class SearchWhere you enrol
- Course outlinesSyllabi from past terms (UW login)
- UW FlowStudent ratings and reviews
- WClassesPast sections back to 2004, and who taught them
Calendar versions
Graduate Calendar
- Winter 2026
- Spring 2026
- Fall 2026 (shown here)
- Winter 2024 and earlier in the calendar archives