Machine learning
Statistical algorithms that learn from data without explicit programming.
Machine learning is a field of study in artificial intelligence concerned with the development and study of statistical algorithms that can learn from data and generalize to unseen data, thus performing tasks without being explicitly programmed. Advances in deep learning have allowed neural networks to surpass many previous machine learning approaches in performance. The field has roots in decades of efforts to study human cognitive processes, with foundational contributions from researchers such as Donald Hebb, Walter Pitts, and Warren McCulloch.
- field
- Artificial intelligence, statistics, computer science
- known_for
- Statistical algorithms that learn from data; deep learning; empirical risk minimization
- foundational_concept
- Probably approximately correct learning
- early_program
- Checkers-playing program by Arthur Samuel (1950s)
Lore & Background
The earliest machine learning program was introduced in the 1950s, when Samuel invented a computer program that calculated the chance of winning in checkers for each side.
Reader's Guide
Machine learning grew out of the quest for artificial intelligence, but by the 1980s, an increasing emphasis on logical, knowledge-based approaches caused a rift between AI and machine learning. Neural network research was abandoned by AI and computer science around that time, but was continued by researchers from other disciplines, including John Hopfield, David Rumelhart, and Geoffrey Hinton. Machine learning, reorganised as its own field, started to flourish in the 1990s, shifting focus from achieving artificial intelligence to tackling solvable practical problems, and borrowing methods from statistics, fuzzy logic, and probability theory. Machine learning and data mining often employ the same methods but with different goals: machine learning focuses on prediction based on known properties, while data mining focuses on discovering previously unknown properties.
Did You Know?
- The earliest machine learning program was a checkers-playing program introduced in the 1950s.
- In 1960s, Raytheon Company developed Cybertron, an experimental learning machine using reinforcement learning.
- Tom M. Mitchell provided a widely quoted formal definition of machine learning algorithms.
More in Science & Measurement 1-16
Spotted an error? Know more?
This is a living reference — every entry is fact-audited, and reader corrections feed straight into our audit queue. Suggest an edit · See this site's audit record