Category: BayesianPage 3 of 7

Time series, multimodal, mixture model, variational methods

I wish to solve multimodal distribution optimization in two stage approach. Example usecase is newsvendor problem; order quantity should be optimized for the multimodal-shaped demand due to its…

mean field variational bayes adversarial for improvement

no theoretical guarantee for accurate results (Wang18) marginal variances of the parameters are often underestimated (Turner11) Turner11 introduce two problems in applying vEM to time series compactness– separated…

Broderick13_Streaming Variational Bayes

Applying the advantage of Bayesian paradigm(hierarchical modeling, coherent treatment of uncertainty) to big data setting is active in variational Bayes (VB). In VB, marginal likelihood’s variational lower bound…

Hamiltonian and Markov

reference

Huggins20_Validated Variational Inference via Practical Posterior Error Bounds

Three criteria are reviewed. KL divergence -divergence Wasserstein distance bound the difference between expectation of any smooth function. With upper bounds on the function of interest and the…

Approximate Bayesian Inference

Computing the posterior is our target. Expectation to get the marginal as we might not know the exact data distribution. Even if we do, computation is burdensome in…

Diagnosing approximation algorithm with SBC

Regression and other stories

Bayesian inference involves three steps that go beyond classical estimation. 1. data and model are combined to form a posterior distribution, which we typically summarize by a set…

SBC Bootstrap approach

motivation: variability . and heavy computation of point-wise-hyperparmeter SBC check approximate the whole procedure using sensitivity to alpha and may work

automatic differentiation

Use cases: map, mle estimators with gradient descent or quasi Newton methods (1) posterior sampling with HMC (log density, 1 for Euclidean, 3 for Riemannian) standard error, posterior…