STA414H1

UTSG

Statistical Methods for Machine Learning II

Probabilistic foundations of supervised and unsupervised learning methods such as naive Bayes, mixture models, and logistic regression. Gradient-based fitting of composite models including neural nets. Exact inference, stochastic variational inference, and Marko chain Monte Carlo. Variational autoencoders and generative adversarial networks.

View full details on the UofT Academic Calendar
Prereq: STA314H1/ CSC311H1/ CSC311H5/ ( STA314H5, STA315H5)/ CSCC11H3; STA302H1/ STAC67H3/ STA302H5; CSC108H1/ CSC110Y1/ CSC148H1/ CSCA08H3/ CSCA48H3/ CSCA20H3/ CSC108H5/ CSC148H5; ( MAT235H1, MAT236H1)/ MAT235Y1 / MAT237Y1/ MAT257Y1/ ( MATB41H3, MATB42H3)/ ( MAT232H5, MAT236H5)/ ( MAT233H5, MAT236H5); MAT223H1/ MAT224H1/ MAT240H1/ MATA22H3/ MATA23H3/ MAT223H5/ MAT240H5/ MATB24H3/ MAT224H5Breadth: Physical & Mathematical UniversesExcl: CSC412H1, STAD68H3
Easy25%
Useful80%
1
comment
1
rating

Course Info

DepartmentSTA
CampusUTSG (St. George)
Level400
Hours36L
BreadthPhysical & Mathematical Universes
What do you think of STA414H1?

Reviews (1)

From Reddit
Difficulty: 4/5Usefulness: 4/5

CSC412 and its equivalent STA414 is great. Difficult concepts on probabilistic machine learning, heavy on math, but the course material is really interesting and fulfilling. The assignments are fun and relatively easy, exam gets curved. The average was mid 70s.

View on Reddit1 months ago