Revision as of 15:52, 5 February 2022 edit The Anome (talk \| contribs) Edit filter managers, Administrators 260,102 edits Adding short description: "Mathematical methods used in Bayesian inference and machine learning" (Shortdesc helper) ← Previous edit		Revision as of 04:24, 8 February 2022 edit undo Mateen Ulhaq (talk \| contribs) 112 edits →KL divergence: fix: Q from P is consistent with Wikipedia's article on KL Divergence. Tags: Mobile edit Mobile web edit Next edit →
Line 25: === KL divergence === The most common type of variational Bayes uses the [[Kullback–Leibler divergence]] (KL-divergence) of ''PQ'' from ''QP'' as the choice of dissimilarity function. This choice makes this minimization tractable. The KL-divergence is defined as :<math>D_{\mathrm{KL}}(Q \parallel P) \triangleq \sum_\mathbf{Z} Q(\mathbf{Z}) \log \frac{Q(\mathbf{Z})}{P(\mathbf{Z}\mid \mathbf{X})}.</math>

Variational Bayesian methods: Difference between revisions