An explanation of neural networks
In this video, Informa TechTarget editorial director Jen English explains what a neural network is, how it functions and what it might be used for.
Neural networks -- the brains behind AI?
A neural network is a machine learning model that processes data in a way that mimics the functions and structure of the human brain. Falling under the broader field of AI, neural networks contain artificial neurons -- or nodes -- that enable them to carry out tasks like image recognition, predictive modeling, natural language processing and even decision-making.
Neural networks are the underlying technology behind deep learning, which is a subset of machine learning. Here, we'll explain how neural networks operate and their various applications.
Neural networks consist of processors arranged into three layers:
- An input layer, where the raw data first enters the system, similar to how your eyes send visual information to your brain.
- Hidden layers that process the data step by step, with each layer building on the results from the one before it.
- And an outer layer, where the final results come out like a decision or prediction.
These layers work together, passing along information in a way that mimics how neurons in the brain pass signals to each other.
So, how do they actually learn? Neural networks are trained on large amounts of data on the topic they will be used for. Training consists of providing inputs, telling the network what the output should be and a set of rules to guide their thought process, so to speak.
For example, to develop a system that identifies different types of vehicles, the training dataset might include images of cars, trucks, bicycles and random objects like furniture or animals. Each image comes with a label, such as "car," "truck" or "not a vehicle." By using these labels, the system refines its internal weightings, enhancing its ability to accurately recognize and classify vehicles in future inputs.
That's a basic example, but neural networks are used in a wide variety of industries. Some notable applications include:
- Speech-to-text transcription.
- Handwriting recognition.
- Process and quality control.
- Personal assistants.
- Chatbots.
- Stock market prediction.
- Delivery route planning and optimization.
- And so much more.
Does your organization use neural networks? How so? Share your experience in the comments, and remember to like and subscribe, too.
Tommy Everson is an assistant editor for video content at TechTarget. He assists in content creation for TechTarget's YouTube channel and TikTok page.
Sabrina Polin is a managing editor of video content for the Learning Content team. She plans and develops video content for TechTarget's editorial YouTube channel, Eye on Tech. Previously, Sabrina was a reporter for the Products Content team.