You often hear AI, machine learning and deep learning used together. They are related like nesting dolls: deep learning is a type of machine learning, which is a type of AI.
Artificial intelligence
The broad field of making machines that can perform tasks that need human-like intelligence.
Machine learning (ML)
A branch of AI where computers learn from data instead of following only hand-written rules. The system finds patterns in examples and uses them to make predictions.
Example: a bank feeds an ML model thousands of past loan records labelled “repaid” or “not repaid”. The model learns which factors predict repayment and scores new applications.
Three main types of machine learning
- Supervised learning: learns from labelled examples, such as emails marked “spam” or “not spam”.
- Unsupervised learning: finds hidden groups in unlabelled data, such as grouping customers by buying habits.
- Reinforcement learning: learns by trial and error with rewards, like an AI learning to play a game by winning and losing.
Deep learning (DL)
Deep learning uses artificial neural networks, loosely inspired by the brain, with many layers (“deep”) of connected nodes. Each layer learns increasingly complex features. In image recognition, early layers detect edges, middle layers detect shapes, and later layers recognise whole objects like faces.
Key differences
- Data needs: traditional ML can work with smaller datasets; deep learning usually needs very large amounts of data.
- Features: in traditional ML, humans often choose which features matter; deep learning figures out features itself.
- Computing power: deep learning needs powerful hardware such as GPUs.
- Best for: traditional ML suits structured data like spreadsheets; deep learning excels at images, speech and language.
Where deep learning shines
Voice assistants, face recognition, self-driving car vision, medical image analysis, translation and modern AI chatbots all rely on deep learning.
