Epistemic uncertainty quantification in deep learning classification by the Delta method.

导出 LinkOut

翻译打印收藏纠错

作者： Geir K,Nilsen ^[1] ; Antonella Z,Munthe-Kaas ^[2] ; Hans J,Skaug ^[3] ; Morten,Brun ^[4]

作者单位： Department of Mathematics, University of Bergen, Norway. Electronic address: geir.kjetil.nilsen@gmail.com. ^[1] Department of Mathematics, University of Bergen, Norway. Electronic address: antonella.zanna@uib.no. ^[2] Department of Mathematics, University of Bergen, Norway. Electronic address: hans.skaug@uib.no. ^[3] Department of Mathematics, University of Bergen, Norway. Electronic address: morten.brun@uib.no. ^[4]

关键词 Deep learning Fisher information Hessian Neural networks Predictive epistemic uncertainty Uncertainty quantification

主题词模型, 统计学(Models, Statistical);概率(Probability);偏差(Uncertainty)

DOI 10.1016/j.neunet.2021.10.014

PMID 34749029

发布时间 2021-12-14

翻译满意度评价：

准确一般字义偏差语意偏离译文作用小提交