Machine learning is a collection of algorithms that learn from data. Supervised learning: learning from labelled data (spam/not spam) — classification and regression. Unsupervised learning: finding hidden structures in unlabelled data — clustering, dimensionality reduction.
Reinforcement learning: an agent acts in an environment and learns through reward/punishment — AlphaGo and robotics. The overfitting problem: the model "memorises" training data and performs poorly on new data. Train/validation/test split and cross-validation solve this problem. Scikit-learn (Python) is the easiest library for getting started with ML.
Our courses on this topic