BriefGPT.xyz
Aug, 2020
神经因果表示的摊销学习
Amortized learning of neural causal representations
HTML
PDF
Nan Rosemary Ke, Jane. X. Wang, Jovana Mitrovic, Martin Szummer, Danilo J. Rezende
TL;DR
本文提出了一种名为因果关系网络的算法,采用神经网络学习因果模型,并使用连续表示方法表示因果模型,从而更好地处理大量变量和利用先前的知识帮助学习新的因果模型,同时提出一种基于解码的评估指标。在合成数据的测试中取得了高精度和快速适应新因果模型的效果。
Abstract
causal models
can compactly and efficiently encode the data-generating process under all interventions and hence may generalize better under changes in distribution. These models are often represented as
bayesian networ
→