ML

Artificial Intelligence (AI) is the broad discipline of building machines that exhibit intelligent behaviour. Machine Learning (ML) is a subset of AI that achieves intelligence through data-driven learning rather than explicit rule programming. All machine learning is AI, but not all AI is machine learning — rule-based expert systems, for example, are AI but not ML.

Deep Learning (DL) is a subfield of Machine Learning that uses multi-layer artificial neural networks. All deep learning is machine learning, but not all machine learning is deep learning — classical ML also includes decision trees, random forests, SVMs, and linear models, which do not use deep neural networks. Deep learning tends to outperform classical ML when large datasets and raw inputs (images, audio, text) are involved.

The data requirements depend on the learning paradigm. Supervised ML requires labelled examples — input–output pairs where a human has provided the correct answer. Unsupervised ML works on unlabelled raw data. Reinforcement learning requires an environment that provides reward signals in response to actions. In general, ML models improve with more data, though the marginal benefit decreases, and data quality matters more than raw quantity.