BriefGPT.xyz
Jun, 2018
结构化 Horseshoe 先验贝叶斯神经网络的变分学习
Structured Variational Learning of Bayesian Neural Networks with Horseshoe Priors
HTML
PDF
Soumya Ghosh, Jiayu Yao, Finale Doshi-Velez
TL;DR
本文介绍了针对贝叶斯神经网络的节点选取问题，使用马蹄铁Prior激活Bayesian神经网络的方法，提高了模型紧凑度，不影响性能表现，特别适用于小样本情况下的强化学习。
Abstract
bayesian neural networks
(BNNs) have recently received increasing attention for their ability to provide well-calibrated posterior uncertainties. However,
model selection
---even choosing the number of nodes---rem
→