Programming a Neural Network Tutorial: Backward Propagation

Author(s): Adam Ross Nelson

Originally published on Towards AI.

Densely connected artificial neural network learning processes

This member-only story is on us. Upgrade to access all of Medium.

To fully understand the theory of how neural networks learn you have to have a close look at backward propagation. This tutorial provides that close look.

Following the previous tutorial in this series which examined the math underlying neural networks, this tutorial offers a closer look at the formal learning (or training) process necessary for neural networks to produce meaningful predictions.

More succinctly, the previous lesson taught about the general structure of artificial neural networks (ANNs) and how data flows through the model, and how each layer processes the data. You can find the previous lesson which includes coding examples in Python and lesson here.

Studying dense neural networks are an important step on the journey of learning data science. Dense neural networks are used in various domains to solve regression and classification problems.

Image Credit: An artistic rendering of a machine learning loss landscape. Author’s illustration created in Canva with images from Dall-E.

While other more traditional and classic approaches can also solve regression and classification problems, we turn to neural networks, which can often pick up more complex predictive signals from the data. Also neural networks are also in more complex deep… Read the full blog for free on Medium.

Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.

Published via Towards AI