Duke Data Science Educational Learning Experiences

Below are events announced from the vice provost for research. Register for both events on the Data Science website: https://plus.datascience.duke.edu/learn-ds

Implementing Generative Adversarial Networks in TensorFlow

Monday, March 18 from 4:00–6:00 PM in The Edge Workshop (Bostock Library first floor)

• Instructor: Kevin Liang
• Generative Adversarial Networks (GANs) have drawn considerable interest in the deep learning community for their ability to synthesize realistic examples across a wide range of domains. Example applications include generating made-up celebrity faces, envisioning how Van Gogh would have painted the Duke Chapel, cracking Vigenère ciphers, and spawning fake news. In this session, we'll review the GAN conceptual framework and then build a simple GAN together in TensorFlow, Google's deep learning library. Bring a laptop and be ready to code!

Image Classification in Tensorflow

Monday, March 25 from 4:00–6:00 PM in The Edge Workshop (Bostock Library first floor)

• Instructor: John Sigman
• Convolutional Neural Networks (CNNs) are the backbone of modern computer vision techniques. In this session, we will see how they are implemented in Tensorflow, a popular Deep Learning framework. Together we will code, train, and evaluate an image classification network. Understanding how to use Tensorflow on this simple problem will help you to use more complicated models.