1.How can we perform efficient inference and learning in directed probabilistic models, in the presence of continuous latent variables with intractable posterior distributions, and large datasets?
keyword intractable posterior distributions
reference:https://www.cs.toronto.edu/~rsalakhu/talks/talk_inference.pdf
学单词系列,我寻思这论文intractable能表达的意思比hard要多多少?
这就是一个简单的贝叶斯推断,然而这个贝叶斯推断需要后验概率,这个在现实世界是很难得到的。
We introduce a stochastic variational inference and learning algorithm that scales to large datasets and, under some mild differentiabili