Deep learning
What is Deep learning?
A type of machine learning built on neural networks with many layers. The extra layers let the system learn complex patterns straight from raw data such as images, sound or text, with no person defining the features first. It's what made the current wave of AI, including language models, possible.
Why it matters
Deep learning is the kind of machine learning behind the current wave of AI, language models included. It uses neural networks with many layers, and the depth is the point. The extra layers let the system learn complicated patterns straight from raw data such as images, sound or text, without a person defining in advance what to look for. That is what made the leap possible. Older methods needed experts to hand-pick the features that mattered; deep learning works them out on its own, given enough data and computing power. For a business, it explains why AI capability jumped so sharply once data and hardware caught up.
In practice
A hospital trains a deep-learning model on labelled medical images and it learns to spot signs of disease that would take a specialist to see, with no one telling it which pixels count. Speech recognition, translation and the models behind today’s AI assistants all rest on the same approach. The trade-off is appetite. Deep learning needs a lot of data and serious hardware, which favours organisations that have both.
