An adaptive mixture-population Monte Carlo method for likelihood-free inference

12/01/2021
by   Zhijian He, et al.
0

This paper focuses on variational inference with intractable likelihood functions that can be unbiasedly estimated. A flexible variational approximation based on Gaussian mixtures is developed, by adopting the mixture population Monte Carlo (MPMC) algorithm in <cit.>. MPMC updates iteratively the parameters of mixture distributions with importance sampling computations, instead of the complicated gradient estimation of the optimization objective in usual variational Bayes. Noticing that MPMC uses a fixed number of mixture components, which is difficult to predict for real applications, we further propose an automatic component–updating procedure to derive an appropriate number of components. The derived adaptive MPMC algorithm is capable of finding good approximations of the multi-modal posterior distributions even with a standard Gaussian as the initial distribution, as demonstrated in our numerical experiments.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
02/23/2022

Cyclical Variational Bayes Monte Carlo for Efficient Multi-Modal Posterior Distributions Evaluation

Statistical model updating is frequently used in engineering to calculat...
research
09/24/2018

Flexible Mixture Modeling on Constrained Spaces

This paper addresses challenges in flexibly modeling multimodal data tha...
research
09/23/2022

A Unified Perspective on Natural Gradient Variational Inference with Gaussian Mixture Models

Variational inference with Gaussian mixture models (GMMs) enables learni...
research
09/30/2022

Learning with MISELBO: The Mixture Cookbook

Mixture models in variational inference (VI) is an active field of resea...
research
07/09/2019

Decentralized Gaussian Mixture Fusion through Unified Quotient Approximations

This work examines the problem of using finite Gaussian mixtures (GM) pr...
research
11/17/2020

Recursive Inference for Variational Autoencoders

Inference networks of traditional Variational Autoencoders (VAEs) are ty...
research
03/13/2019

Generalized Elliptical Slice Sampling with Regional Pseudo-priors

In this paper, we propose a MCMC algorithm based on elliptical slice sam...

Please sign up or login with your details

Forgot password? Click here to reset