Bayesian Methods
4concepts
31flashcards
31minutes of reading
- 01 Priors, Conjugacy and the Posterior How a prior functions as pseudo-data, why conjugate families make the posterior a closed-form update, and why "uninformative" priors are informative on some scale.
- 02 Gaussian Processes Placing a prior over functions rather than parameters, which gives exact posterior uncertainty that grows away from the data, at a cubic cost that determines where they are usable.
- 03 MCMC and Hamiltonian Monte Carlo Why sampling from an unnormalised posterior is possible at all, why random-walk proposals fail in high dimensions, and how using gradient information turns a random walk into directed motion.
- 04 Variational Inference and the ELBO Turning integration into optimisation by fitting a tractable distribution to the posterior, and the specific bias that comes from minimising the reverse KL divergence.