Introduction to Neural Networks with Tensorflow

12:15 PM - 12:40 PM on July 16, 2016, Room CR4

Mike Craig

Audience level:

intermediate

Watch:

https://www.youtube.com/watch?v=Qqhs4w57sFQ

Description

Neural networks are a powerful machine learning model that are used in many applications, from computer vision and speech recognition to time series modeling and natural language processing. TensorFlow is an open source machine learning library developed by Google that can be used to easily create these types of models. I will give an overview of neural networks, and show how to implement and use these models using TensorFlow's Python API.

Abstract

Neural networks are a general purpose machine learning model that can approximate any function. Understanding their implementation and use is an important skill for both the developer and the data scientist.

There are many machine learning libraries out there. TensorFlow, an open source project from Google, is one such library that allows developers to build and design large scale network models. It allows the developer full control over the model's structure while hiding tedious implementation details. In this talk, I will give an introduction to the properties and structure of a neural network model, and use TensorFlow to easily build such models.