BriefGPT.xyz
Jun, 2021
解开筛选,数据增强和先验在冷后验效应中的作用
Disentangling the Roles of Curation, Data-Augmentation and the Prior in the Cold Posterior Effect
HTML
PDF
Lorenzo Noci, Kevin Roth, Gregor Bachmann, Sebastian Nowozin, Thomas Hofmann
TL;DR
研究表明Bayesian深度学习中的冷链效应是一个多因素的问题,直接利用温度参数、数据增强或先验偏差可能并不能解决该问题。
Abstract
The "
cold posterior effect
" (CPE) in Bayesian deep learning describes the uncomforting observation that the predictive performance of
bayesian neural networks
can be significantly improved if the Bayes posterior
→