张量化流：一种变分推断工具（英文版）

英文标题：Tensorizing flows: a tool for variational inference中文摘要：本文提出了一种引入张量网络构造的参考分布的扩展归一化流模型，将其应用于困难的变分推断任务中，显示出优于单独使用流或张量网络的结果。英文摘要：Fueled by the expressive power of deep neural networks, normalizing flowshave achieved spectacular success in generative modeling, or learning to drawnew samples from a distribution given a finite dataset of training samples.Normalizing flows have also been applied successfully to variational inference,wherein one attempts to learn a sampler based on an expression fo