useR! 2020: Deep Learning with Keras and TensorFlow (S. Elsinghorst), tutorial

Dr. Shirin Elsinghorst presents on using Keras and TensorFlow. Dr. Elsinghorst begins the presentation with background information on herself as well as goals for the workshop. The tutorial begins with the basics of deep learning, what cross-entropy and loss is, follows with a description of activation functions, and additional resources. Dr. Elsinghorst then digs into an in depth demonstration covering building deep neural networks with Keras and Tensorflow. Visualizing models with TensorBoard, and making predictions on test data. At the end of the tutorial there is a questions and answers session. This video is part of the virtual useR! 2020 conference. Find supplementary material on our website https://user2020.r-project.org/. Main Sections *Does not begin speaking until 7:38 7:38 Introduction 11:01 About Me 11:54 Workshop agenda 12:50 Neutral networks 16:35 Activation function 20:25 Softmax, cross-entropy, back propagation and gradient descent optimization 30:47 Deep learning with keras & tensorflow 35:51 Keras API’s 38:28 Practical part: convolutional neural networks 48:27 Additional resources 49:02 Demonstration using Rstudio 1:49:24 More resources 1:53:30 Questions 2:06:10 Wrap up More Resources Main Site: https://www.r-consortium.org/ News: https://www.r-consortium.org/news Blog: https://www.r-consortium.org/news/blog Join: https://www.r-consortium.org/about/join Twitter: https://twitter.com/Rconsortium LinkedIn: https://www.linkedin.com/company/r-consortium/