Variational Inference (also known as Variational Bayes) is a method to find an approximation to a posterior distribution, by casting it as an optimization problem.
In this course, we will learn everything needed for understanding and do VI in R.
Note: with this course you also get access to the Latent Dirichlet Allocation (LDA) mini-course.
Curriculum
- 11 Sections
- 32 Lessons
- Lifetime
Expand all sectionsCollapse all sections
-
Administration1
-
Intro3
-
KL Divergence5
-
VI (using KL) vs. ML1
-
ELBO & "Mean Field"2
-
Coordinate Ascent VI (CAVI)7
-
Exponential Family3
-
VI vs. EM1
-
Stochastic VI / Advanced VI6
-
Expectation Propagation2
-
Variational Auto-Encoder1
Requirements
- Linear Algebra
- Calculus
- Intro to Probability
- Intro to Statistics
- Bayesian Inference
- R Programming Language