|
Home
| Subject Search
| Help
| Symbols Help
| Pre-Reg Help
| Final Exam Schedule
| My Selections
|
Searched for: 1 subject found.
6.7970[J] Symmetry and its Applications to Machine Learning
(New)(
)
(Same subject as 8.750[J])
Prereq: 6.100A, 6.1210, and 18.C06
Units: 3-0-9
Lecture: MW2.30-4 (32-141)
Introduces group representation theory to design symmetry-preserving machine learning algorithms, emphasizing the connections between mathematics, physics, and data-driven models. Students implement core mathematical concepts in code to construct algorithms that operate on structured data — such as graphs, geometric objects, and scientific datasets — while preserving their underlying symmetries. Topics include finite and infinite groups (with an introduction to Lie algebras), various group representations (regular, reducible, and irreducible), tensor products and decompositions, Fourier analysis and convolutions, statistics and sampling of representation vector spaces, and symmetry-breaking mechanisms. Previous knowledge of group theory is not required but is beneficial.
T. Smidt
No textbook information available
| ||||||||
