Bayesian Machine Learning

This is materials on Bayesian Machine Learning that I developed during autumns in 2018, 2019, 2020. Despite that it was three times, mistakes and misprints are presented almost surely. I would like to thank Anna Kuzina and Ruslan Kostoev for their thoughtful comments and help.

For each class there is some combination of following materials:

• slides: contain some theory, sometimes derivations for particular examples
• notes: contain derivations of the discussed topic for particular model or exp. family case
• notebooks: contain illustrative experiment or implementation of the model of the discussed topic
• whiteboard\& video: zoom-record of actual seminar and live-made notes

#	Topic	Materials	Whiteboard	Video
1	MaxEnt, Exponential Families Mean and Natural Parameters	slides, notebook	whiteboard	TBA
2	Bayesian Linear Regression: RVM and Sequential Updates	slides, notebooks RVM, notebooks RVM	None	TBA
3	Conjugate Priors	slides	whiteboard	TBA
4,5	Approximation Inference for Special Non-Conjugate Models	slides	whiteboard4, whiteboard5	TBA
6	EM Algorithm (and Exp. Families)	slides, notes for exp. families	whiteboard	TBA
7	Mean Field (and Exp. Families)	slides, derivations for simple models and exp. families	whiteboard	TBA
8	Reinforce and Reparametrization	slides, notebook	whiteboard	TBA
12	Variational Dropout	notebook	whiteboard	TBA
14	Normalizing Flows	slides, notebook	None	TBA
HW	EM Mixture for Robust PCA derivation	notes	None	TBA
18	MCMC	notebook	None	TBA

I suppose most visitors to this page are in seek of materials to learn on Bayesian machine learning. Hence, I could not avoid the recommendation of the Deep Bayes materials.