Anomaly Detection with Variational Autoencoders

Hosted by Colorado Data Science

Public group


**Make sure to register for this event on Eventbrite (**

Join us on Tuesday, August 8th at Galvanize Platte Street for a talk on Anomaly Detection with Variational Autoencoders from Eric Harper from NVIDIA.

About the talk

NVIDIA GPUs are powering virtual reality, high performance computing, artificial intelligence, and big data. Deep learning models are responsible for some of the most impactful technologies today including autonomous driving, machine translation, and AI assistants. In the first part of this talk we will give an overview of deep learning, GPU accelerated analytics, and how enterprises are solving new business problems with these technologies. In the second part of the talk we will hone in on a (relatively) new deep learning model: variational autoencoders. Variational autoencoders are being used for generative modeling and anomaly detection. In addition to achieving higher accuracy on many anomaly detection tasks, variational autoencoders also have a sound mathematical background which may prove useful in industries that are heavily regulated, like finance. If there’s time, we’ll give a live demo.

About the speaker

Eric Harper - NVIDIA

Solutions Architect, Deep Learning and AI

Eric is a Data Scientist and Solutions Architect for NVIDIA ( with a focus on deep learning for enterprise. Before NVIDIA he was the Lead Data Scientist for DISH Media Sales and before entering industry he was a Postdoctoral Researcher in Low-Dimensional Topology at McMaster University and the University of Quebec at Montreal.

**Make sure to register for this event on Eventbrite (**