Multiplicative Normalizing Flows for Variational Bayesian Neural Networks

2017-03-06 · ICML 2017 8 · Christos Louizos, Max Welling

We reinterpret multiplicative noise in neural networks as auxiliary random variables that augment the approximate posterior in a variational setting for Bayesian neural networks. We show that through this interpretation it is both efficient and straightforward to improve the approximation by employing normalizing flows while still allowing for local reparametrizations and a tractable lower bound. In experiments we show that with this new approximation we can significantly improve upon classical mean field for Bayesian neural networks on both predictive accuracy as well as predictive uncertainty.

📄 PDF Abstract BibTeX arXiv:1703.01961

Code (7)

AMLab-Amsterdam/MNF_VBNN tf

VietHoang1512/MNF tf

akashrajkn/waffles-and-posteriors tf

beauCoker/bayesian_neural_networks pytorch

iclr-mcbn/mcbn tf

icml-mcbn/mcbn tf

sliorde/generating-neural-networks-with-neural-networks tf

Randomized Value Functions via Multiplicative Normalizing Flows

2018-06-06 · Ahmed Touati, Harsh Satija, Joshua Romoff, Joelle Pineau 외

Randomized value functions offer a promising approach towards the challenge of efficient exploration in complex environments with high dimensional state and action spaces. Unlike traditional point estimate methods, rando…

Efficient Exploration Thompson Sampling

Convolutional Normalizing Flows

2017-11-07 · ICLR 2018 1 · Guoqing Zheng, Yiming Yang, Jaime Carbonell

Bayesian posterior inference is prevalent in various machine learning problems. Variational inference provides one way to approximate the posterior distribution, however its expressive power is limited and so is the accu…

Variational Inference

Amortized variational transdimensional inference

2025-06-05 · Laurence Davies, Dan MacKinlay, Rafael Oliveira, Scott A. Sisson

The expressiveness of flow-based models combined with stochastic variational inference (SVI) has, in recent years, expanded the application of optimization-based Bayesian inference to include problems with complex data r…

Bayesian Inference Bayesian Optimization Variational Inference

Sylvester Normalizing Flows for Variational Inference

2018-03-15 · Rianne van den Berg, Leonard Hasenclever, Jakub M. Tomczak, Max Welling

Variational inference relies on flexible approximate posterior distributions. Normalizing flows provide a general recipe to construct flexible variational posteriors. We introduce Sylvester normalizing flows, which can b…

Variational Inference

Improved Variational Bayesian Phylogenetic Inference with Normalizing Flows

2020-12-01 · NeurIPS 2020 12 · Cheng Zhang

Variational Bayesian phylogenetic inference (VBPI) provides a promising general variational framework for efficient estimation of phylogenetic posteriors. However, the current diagonal Lognormal branch length approximati…