Visualizing autoencoders - Made with TensorFlow.js

Meet Doug Duhaime, a software engineer and professor at Yale University who has been exploring autoencoders in TensorFlow.js. See how he creates impressive transformations of high dimensional space to 2D space, or provides demonstrations to envision the mechanics of machine learning to educate students around dimensionality reduction. Doug has been using TensorFlow.js to demonstrate what autoencoders are able to learn after being trained on classic data sets such as MNIST and Celeb-A to walk the learnt 2 dimensional latent space and visualize new faces never seen before all in realtime in the browser. Try it for yourself: Autoencoders by Douglas Duhaime → https://goo.gle/2XvSlkE Want to be on the show? Use #MadeWithTFJS to share your own creations on social media and we may feature you in our next show. Catch more #MadeWithTFJS interviews → http://goo.gle/made-with-tfjs Subscribe to the TensorFlow channel → https://goo.gle/TensorFlow product: TensorFlow - TensorFlow JS, TensorFlow - General; fullname: Jason Mayes;