00:17 Roadmap 00:56 Deep Learning 02:19 Drawbacks of Standard Deep Learning 03:13 Probabilistic Machine Learning 04:17 Bayesian Neural Networks 07:28 History of Bayesian Neural Networks 08:15 Modern Revival: Bayesian Deep Learning 08:47 Probabilistic Programming with Edward 11:14 Edward 14:53 Inference in Edward 15:54 Variational Inference 18:24 Black Box Variational Inference with Edward 18:50 Dropout as a Bayesian Approximation 22:34 MC Dropout Experiments 22:54 Experiments 27:20 Model Specific versus Black Box 29:29 Current Research in Variational Inference 30:27 QnA