We describe a hierarchical Bayesian approach for inference about a parameter $θ$ lower-bounded by $α$ with uncertain $α$, derive some basic identities for posterior analysis about $(θ,α)$, and provide illustrations for normal and Poisson models. For the normal case with unknown mean $θ$ and known variance $σ^2$, we obtain Bayes estimators of $θ$ that take values on $\mathbb{R}$, but that are equally adapted to a lower-bound constraint in being minimax under squared error loss for the constrained problem.
Uncertainty Estimations by Softplus normalization in Bayesian Convolutional Neural Networks with Variational Inference
TU Kaiserslautern(凯泽斯劳滕工业大学);MindGarage;Imperial College London(帝国理工学院);NeuralSpace;Luleå University of Technology(吕勒奥理工大学)
我们介绍了一种新的不确定性估计贝叶斯卷积神经网络与变分推理的分类任务。通过在最后一层对Softplus函数的输出进行归一化,我们以连贯的方式估计任意和认知的不确定性。权重的后验概率分布由Bayes by Backprop推断。首先,我们展示了这种可靠的变分推理方法可以作为各种网络架构的基本构造。在监督学习设置(MNIST,CIFAR-10,CIFAR-100)的多个数据集上,这种变分推理方法在相同的体系结构中实现了与频率论推理相当的性能,而两个必要的,不确定性和正则化的度量自然地结合在一起。其次,我们研究了我们提出的任意性和认知不确定性的度量是如何推导出来的,并在上述数据集上进行了验证。
Uncertainty-Aware Surrogate-based Amortized Bayesian Inference for Computationally Expensive Models
Department of Stochastic Simulation and Safety Research for Hydrosystems(随机模拟与安全研究部门);Cluster of Excellence SimTech(卓越中心SimTech);University of Stuttgart(斯图加特大学);Department of Statistics(统计学系);TU Dortmund University(多特蒙德技术大学)
Harvard T.H. Chan School of Public Health(哈佛陈曾熙公共卫生学院);Beth Israel Deaconess Medical Center(贝斯以色列女执事医疗中心);The Lundquist Institute for Biomedical Innovation at Harbor-UCLA Medical Center(哈珀-加州大学洛杉矶分校医学中心伦德奎斯特生物医学创新研究所)
Online Statistical Inference of Constrained Stochastic Optimization via Random Scaling
Department of Statistics, The University of Chicago(芝加哥大学统计系);Department of Statistics, University of California, Irvine(加州大学尔湾分校统计系);School of Industrial and Systems Engineering, Georgia Institute of Technology(佐治亚理工学院工业与系统工程学院)