Deep Learning for Beginners

Description: In the past few years, there has been a significant breakthrough in the area of machine learning. In this course we will introduce the state-of-the-art machine learning method, called deep learning. In the first part of the course, we will present related concepts and the working principle of deep learning. Then in the second part, we will show how to use deep learning to solve a real-world problem --- recognition of handwritten digits. Audience will learn to write their own code to solve the problem. This hands-on experience will let them understand deep learning inside out. Then, we move on to use one of the deep learning libraries (Tensorflow) to tackle exactly the same problem. We will show how to improve the recognition rate by designing a better neural network. In the end, we will discuss other applications of deep learning.

Instructor: Weiguang Guan, SHARCNET, McMaster University.

Prerequisites: It will help if you know basics of Python and a little math (calculus, linear algebra)

Course materials:
See the link: http://guanw.sharcnet.ca/.