What is deep learning used for?#
Instead of being programmed with rules, deep learning systems automatically discover complex patterns and features directly from data, enabling them to make highly accurate predictions or decisions about new, unseen information.
How does deep learning work?#
Deep learning uses a structure called a neural network, which is made up of many layers of interconnected digital "neurons." When presented with data, like an image or a sound file, each layer processes the information it receives and passes a more refined output to the next layer. The initial layers might identify very simple patterns like colours or edges, while deeper layers combine these patterns to recognise more complex concepts like faces, objects, or words.
During a "training" phase, the system is shown millions of examples (e.g., images labelled "cat" or financial transactions labelled "fraudulent"). It constantly adjusts the connections between its neurons to improve the accuracy of its predictions. This process allows it to learn without being explicitly told which features to look for, making it highly effective for unstructured data like images, text, and audio.
How is deep learning different from machine learning?#
Deep learning is an advanced form of machine learning. While traditional machine learning often requires a human expert to manually select the most important features from data for the system to analyse, a deep learning system can automatically discover those relevant features on its own. This ability to learn from raw data makes it more powerful and scalable for solving extremely complex problems.