The variational approximation for Bayesian inference

991 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Using labeled data to evaluate change detectors in a multivariate streaming environment (2009, neutral): “There are two main problems for real-time and accurate semantic segmentation of river ice: (1) The segmentation accuracies of some semantic segmentation networks are very high, but due to their very complex model structures and some lightweight terminals often do not have the computing power of the…”
A Novel Adaptive Kalman Filter With Inaccurate Process and Measurement Noise Covariance Matrices (2018, methodology): “, [16], [17] p(xk,Pk|k−1,Rk|z1:k) ≈ q(xk)q(Pk|k−1)q(Rk) (17) where q(·) represents the approximate posterior PDF of p(·), and q(xk), q(Pk|k−1) and q(Rk) are given by minimizing the KullbackLeibler divergence (KLD) between the factored approximate posterior PDF q(xk)q(Pk|k−1)q(Rk) and true joint post”
VIREL: A Variational Inference Framework for Reinforcement Learning (2018, methodology): “onstant. However, as the E-step is typically intractable, we resort to variational inference, which approximates difﬁcult posteriors using a parameterised variational distribution q(h;) (Beal, 2003; Tzikas et al., 2008; Fox and Roberts, 2010). The objective of variational inference is therefore to r”
A Novel Robust Student's t-Based Kalman Filter (2017, neutral): “[25], [26] {q(xk), q(ξk), q(Σk), q(λk)} = argminKLD (q(xk)q(ξk)q(Σk)q(λk)||p(xk, ξk,Σk, λk|z1:k)), (33) where KLD(q(x)||p(x)) , ∫ q(x) log q(x) p(x)dx is the KLD.”