Current browse context:
stat.ML
Change to browse by:
References & Citations
Statistics > Machine Learning
Title: Kernel Implicit Variational Inference
(Submitted on 29 May 2017 (v1), last revised 23 Feb 2018 (this version, v3))
Abstract: Recent progress in variational inference has paid much attention to the flexibility of variational posteriors. One promising direction is to use implicit distributions, i.e., distributions without tractable densities as the variational posterior. However, existing methods on implicit posteriors still face challenges of noisy estimation and computational infeasibility when applied to models with high-dimensional latent variables. In this paper, we present a new approach named Kernel Implicit Variational Inference that addresses these challenges. As far as we know, for the first time implicit variational inference is successfully applied to Bayesian neural networks, which shows promising results on both regression and classification tasks.
Submission history
From: Jiaxin Shi [view email][v1] Mon, 29 May 2017 11:11:35 GMT (922kb,D)
[v2] Thu, 22 Feb 2018 13:49:00 GMT (8304kb,D)
[v3] Fri, 23 Feb 2018 15:45:59 GMT (8304kb,D)
Link back to: arXiv, form interface, contact.