paper-with-me

Papers

Forward $χ^2$ Divergence Based Variational Importance Sampling

2023-11-04 · Chengrui Li, Yule Wang, Weihan Li, Anqi Wu

Maximizing the log-likelihood is a crucial aspect of learning latent variable models, and variational inference (VI) stands as the commonly adopted method. However, VI can encounter challenges in achieving a high log-likelihood when dealing with complicated posterior distributions. In response to this limitation, we introduce a novel variational importance sampling (VIS) approach that directly estimates and maximizes the log-likelihood. VIS leverages the optimal proposal distribution, achieved by minimizing the forward $\chi^2$ divergence, to enhance log-likelihood estimation. We apply VIS to various popular latent variable models, including mixture models, variational auto-encoders, and partially observable generalized linear models. Results demonstrate that our approach consistently outperforms state-of-the-art baselines, both in terms of log-likelihood and model parameter estimation.

📄 PDF Abstract BibTeX arXiv:2311.02516

Code (0)

등록된 구현이 없습니다.

Tasks

parameter estimationVariational Inference

Methods 이 논문이 사용한 방법론

Variational Inference 설명 없음

Similar Papers 제목 키워드 기반

Variational Refinement for Importance SamplingUsing the Forward Kullback-Leibler Divergence

2020-11-23 · pproximateinference AABI Symposium 2021 1 · Ghassen Jerfel, Serena Lutong Wang, Clara Fannjiang, Katherine A Heller 외

Variational Inference (VI) is a popular alternative to asymptotically exact sampling in Bayesian inference. Its main workhorse is optimization over a reverse Kullback-Leibler divergence (RKL), which typically underestim…

Bayesian InferenceVariational Inference

Variational Refinement for Importance Sampling Using the Forward Kullback-Leibler Divergence

2021-06-30 · Ghassen Jerfel, Serena Wang, Clara Fannjiang, Katherine A. Heller 외

Variational Inference (VI) is a popular alternative to asymptotically exact sampling in Bayesian inference. Its main workhorse is optimization over a reverse Kullback-Leibler divergence (RKL), which typically underestima…

Bayesian InferenceVariational Inference

Differentiable Annealed Importance Sampling Minimizes The Symmetrized Kullback-Leibler Divergence Between Initial and Target Distribution

2024-05-23 · Johannes Zenn, Robert Bamler

Differentiable annealed importance sampling (DAIS), proposed by Geffner & Domke (2021) and Zhang et al. (2021), allows optimizing over the initial distribution of AIS. In this paper, we show that, in the limit of many tr…

Variational Inference

Nested Variational Inference

2021-06-21 · NeurIPS 2021 12 · Heiko Zimmermann, Hao Wu, Babak Esmaeili, Jan-Willem van de Meent

We develop nested variational inference (NVI), a family of methods that learn proposals for nested importance samplers by minimizing an forward or reverse KL divergence at each level of nesting. NVI is applicable to many…

Variational Inference

A Variational Perspective on Generative Flow Networks

2022-10-14 · Heiko Zimmermann, Fredrik Lindsten, Jan-Willem van de Meent, Christian A. Naesseth

Generative flow networks (GFNs) are a class of models for sequential sampling of composite objects, which approximate a target distribution that is defined in terms of an energy function or a reward. GFNs are typically t…

Variational Inference