British Sign Language Glossaries of Curriculum Terms

BSL AI Glossary - Neural Network definition

Translation: Neural networks contain neurons arranged in layers which are interconnected. Each layer processes the data in a different way. The first layer receives the raw input, such as the pixels of an image. As the data moves through the network, the later layers learn to detect more detailed shapes and features such as ears, or whiskers. The final layer produces the output, for example deciding whether the image shows a dog, a cat, or something else. The learning occurs in the hidden layers, which sit between the input and output layers. Each neuron includes all the neurons from the previous layer, and it multiplies every neuron's weights, adds a bias value, and then passes the result through an activation function. The output of that function is then passed forward to the next layer. Neural networks vary in size and complexity. Some are large and powerful for complex tasks, while others are smaller and faster for simpler tasks.