Objectives On completion of this tutorial, you will be able to:

• Define deep learning and compare it to supervised and unsupervised machine learning
• Identify the key characteristics of artificial neural networks (ANNs) and compare them to the human brain
• List the limitations of ANNs and recall how these are overcome by recurring neural networks (RNNs) and convolutional neural networks (CNNs)
• Define reinforcement learning (RL) and its approach to problem-solving

Tutorial Overview Supervised and unsupervised machine learning (ML) use simple mathematical and statistical tools to process data and produce outputs that can help guide decision-making. However, they are relatively modest tools compared to the range of human intelligence. To develop more complex forms of artificial intelligence (AI), computer scientists and programmers have created artificial neural networks (neural nets) intended to mimic the functioning of the human brain. These neural nets have led to major advances in deep learning and reinforcement learning, enabling the creation of ever-more sophisticated AI tools. This tutorial provides an overview of neural nets, deep learning, and reinforcement learning. Prerequisite Knowledge Unsupervised Machine Learning Tutorial Level: Intermediate Tutorial Duration: 50 minutes