Main resource:

• ‘A First Course in Machine Learning Learning’ - S. Rogers and M. Girolami (Chapter 3 - Bayesian Linear Regression; Chapter 4 - Metropolis-Hastings; Chapter 7 - PPCA and Variational Bayes) - very good introductory textbook!
• ‘Information Theory, Inference, and Learning Algorithms’ - D. MacKay (Chapter 33 -Variational Methods)
• ‘Pattern Recognition and Machine Learning’ - C. Bishop (Chapter 10 - Approximate Inference)

As always, an acompanying video:

• Presentation by David Blei (Columbia University) on foundations and innovations in Variational Infrence - from mean field factorisation approach to black-box variational inference.

Talk will include demonstrations from pymc3, which uses automatic differentiation variational inference (ADVI) and the Nu-U-Turn sampler for MCMC (NUTS).