Machine learning foundations and paradigms

Machine learning definition and purpose

Machine learning (ML) is a field of artificial intelligence focused on developing statistical algorithms that learn from data and generalize to new, unseen data. This allows systems to perform tasks without being explicitly programmed for each specific case. In practice, ML relies on foundations from statistics and mathematical optimization, and many common ML methods can be framed as minimizing an objective (often called empirical risk minimization) to improve performance. The purpose of ML is to enable a learning system to improve its performance on a set of tasks as it gains experience from data. A widely used formal definition describes learning as improving task performance with respect to a performance measure after exposure to experience (training data). ML is also closely tied to the goal of generalization—producing accurate predictions on new examples drawn from an underlying (often unknown) probability distribution. Related areas include data mining (which emphasizes discovering previously unknown properties) and deep learning (a subset of ML that uses neural networks).

0/2 modes complete

Deep learning and neural networks performance

0/2 modes complete