Probabilistic Graphical Models: Principles and Techniques, Daphne Koller and Nir Friedman, 2009 (MIT Press) - This comprehensive textbook is the definitive reference for probabilistic graphical models, detailing exact inference, the necessity of approximate methods, and in-depth coverage of MCMC (including Gibbs sampling in PGMs) and Variational Inference.
Pattern Recognition and Machine Learning, Christopher M. Bishop, 2006 (Springer)DOI: 10.1007/978-0-387-45528-0 - A widely-used textbook that provides a clear and comprehensive introduction to the theoretical foundations of both Markov Chain Monte Carlo (MCMC) and Variational Inference, including the mean-field approximation and ELBO.
Variational Inference: A Review for Statisticians, David M. Blei, Alp Kucukelbir, and Jon D. McAuliffe, 2017Journal of the American Statistical Association, Vol. 112 (American Statistical Association)DOI: 10.1080/01621459.2017.1285773 - This highly influential review article provides an accessible and comprehensive overview of Variational Inference, covering its theoretical underpinnings, the mean-field approximation, and modern advancements like Stochastic Variational Inference.
Bayesian Data Analysis, Andrew Gelman, John B. Carlin, Hal S. Stern, David B. Dunson, Aki Vehtari, and Donald B. Rubin, 2013 (CRC Press)DOI: 10.1201/b16018 - This foundational book on Bayesian statistics offers detailed coverage of Markov Chain Monte Carlo methods, including Gibbs sampling, practical implementation considerations, and convergence diagnostics, essential for robust approximate inference.