BriefGPT.xyz
Feb, 2024
轻量级贝叶斯神经网络具有异方差不确定性的变分推断框架
A Framework for Variational Inference of Lightweight Bayesian Neural Networks with Heteroscedastic Uncertainties
HTML
PDF
David J. Schodt, Ryan Brown, Michael Merritt, Samuel Park, Delsin Menolascino...
TL;DR
本研究提出一种对轻量级贝叶斯神经网络（BNN）进行抽样无关变分推断的简单框架，通过将异方差预测不确定性和经验差异嵌入到学习得到的BNN参数的方差中，提高了预测性能。
Abstract
Obtaining
heteroscedastic predictive uncertainties
from a
bayesian neural network
(BNN) is vital to many applications. Often, heteroscedastic
ale
→