How Neural Networks Learn: Exploring Architecture, Gradient Descent, and Backpropagation
Neural networks drive many artificial intelligence applications today. This course will teach you what’s behind the magic—the dynamics of training neural networks, including backpropagation, gradient descent, and how to optimize network performance.
What you'll learn
So, you understand neural networks conceptually—what they are and generally how they work. But you might still be wondering about all the details that actually make them work.
In this course, How Neural Networks Learn: Exploring Architecture, Gradient Descent, and Backpropagation, you’ll gain an understanding of the details required to build and train a neural network.
First, you’ll explore network architecture—made up of layers, nodes and activation functions—and compare architecture types.
Next, you’ll discover how neural networks adjust and learn to use backpropagation, gradient descent, loss functions, and learning rates.
Finally, you’ll learn how to implement backpropagation and gradient descent using Python.
When you’re finished with this course, you’ll have the skills and knowledge of neural network architectures and learning needed to build and train a neural network.
Table of contents
About the author
Amber has been a software developer and technical trainer for over two decades, sharing her expertise in AI, machine learning, AWS, and Power Apps with students around the world. She has a knack for making complex tech topics easy to grasp, whether you’re a developer, designer, or business professional. Amber holds certifications in machine learning, AWS, and various Microsoft technologies, including her experience as a former Microsoft Certified Trainer. With a strong background in visual com... more