Skip to main content
Welcome to the Advanced Algorithms section of machine learning. This comprehensive guide will help you master neural networks, deep learning algorithms, and build practical machine learning systems.

​
What You Will Learn

In this section, you’ll learn about neural networks (also called deep learning algorithms) as well as decision trees. These are some of the most powerful and widely used machine learning algorithms, and you’ll implement them and get them to work yourself.
This material includes practical advice on building machine learning systems - a unique aspect that will help you make systematic and better decisions about developing practical working applications.

​
Core Topics

1
Neural Networks and Inference
2
Learn how neural networks work and how to carry out inference or prediction. If you download the parameters of a neural network that someone else trained and posted online, using that neural network to make predictions is called inference.
3
Training Your Own Neural Networks
4
Discover how to train your own neural network. If you have a training set of labeled examples (X and Y), you’ll learn how to train the parameters of a neural network yourself.
5
Practical System Development
6
Get practical advice for building machine learning systems. Learn tips that even highly-paid engineers don’t always consistently apply, helping you build systems efficiently and quickly.
7
Decision Trees
8
While decision trees don’t get as much media attention as neural networks, they are widely used and very powerful learning algorithms. There’s a good chance you’ll use them when building applications.

​
Course Structure

The material is organized into four main sections:

Neural Network Layers

Understand how neural network layers work and how data flows from one layer to another

TensorFlow Implementation

Learn to implement neural networks using TensorFlow, one of the leading deep learning frameworks

Vectorization

Master efficient implementations using matrix multiplication for scaling neural networks

Training Networks

Discover how to train neural networks using your own labeled datasets

​
What is Inference?

Inference refers to using a trained neural network to make predictions on new data. When you download pre-trained parameters from the internet and use them to classify images or make predictions, you’re performing inference.

​
Why This Matters

When building a practical machine learning system, you face many decisions:
  • Should you spend more time collecting data?
  • Should you buy a much bigger GPU to build a larger neural network?
  • Which architecture should you choose?
  • How should you optimize training?
With the tips you learn in this section, you’ll be able to make systematic decisions and avoid wasting months on approaches that won’t work well. This practical knowledge separates effective machine learning engineers from those who struggle.

​
Getting Started

Ready to dive in? Start with understanding neural network layers and build your foundation in deep learning. From there, you’ll progress to implementing networks in TensorFlow, optimizing them with vectorization, and finally training your own models. Let’s begin this exciting journey into advanced learning algorithms!

Build docs developers (and LLMs) love

Get started for freeTalk to us