Deep Learning Essentials

IVADO, a scientific and economic data science hub bridging industrial, academic and government partners with expertise in digital intelligence designed the course, and the world-renowned Mila, rallying researchers specialized in Deep Learning, created the content. Mila’s founder and IVADO’s scientific director, Yoshua Bengio, also a professor at Université de Montréal, is a world-leading expert in artificial intelligence and a pioneer in deep learning as well as the scientific director of this course. He is also a joint recipient of the 2018 A.M. Turing Award, “the Nobel Prize of Computing”, for conceptual and engineering breakthroughs that have made deep neural networks a critical component of computing.

Deep Learning is an extension of Machine Learning where machines can learn by experience without human intervention. It is largely influenced by the human brain in the fact that algorithms, or artificial neural networks, are able to learn from massive amounts of data and acquire skills that a human brain would. Thus, Deep learning is now able to tackle a large variety of tasks that were considered out of reach a few years ago in computer vision, signal processing, natural language processing, robotics, and sequential decision-making. Because of these recent advances, various industries are now deploying deep learning models that impact various economic sectors such as transport, health, finance, energy, as well as our daily life in general.

If you are a professional, a scientist or an academic with basic knowledge in mathematics and programming, this MOOC is designed for you! Atop the rich Deep Learning content, discover issues of bias and discrimination in machine learning and benefit from this sociotechnical topic that has proven to be a great eye-opener for many.

What you'll learn

At the end of the MOOC, participants should be able to:

• Understand the basics and terminology related to Deep Learning
• Identify the types of neural networks to use to solve different types of problems
• Get familiar with Deep Learning libraries through practical and tutorial sessions

Syllabus

MODULE 1 Machine Learning (ML) and Experimental Protocol

• Introduction to ML
• ML Tools

MODULE 2 Introduction to Deep Learning

• Modular Approaches
• Backpropagation
• Optimization

MODULE 3 Intro to Convolutional Neural Networks (CNN)

• Introduction to CNN
• CNN Architectures

MODULE 4 Introduction to Recurrent Neural Networks

• Sequence to Sequence Models
• Concepts in Natural Language Processing

MODULE 5 Bias and Discrimination in ML

• Differences of Fairness
• Fairness in Pre- In- and Post-Processing