Machine learning lets computers improve from examples instead of fixed instructions
In short: In 1959, Arthur Samuel described machine learning as a way for computers to improve without being given step-by-step rules. Today the same basic idea sits behind spam filters, recommendation engines and medical tools. Systems build a model from example data and then use it to make predictions about new cases.
Machine learning is a branch of artificial intelligence within computer science. It grew out of the study of pattern recognition and computational learning theory. Rather than following strictly static program instructions, these algorithms build a model from a training set of example observations, then use it to make data-driven predictions or decisions. Before this approach, bioinformatics algorithms had to be written by hand, and tasks such as protein structure prediction proved very difficult that way.
Methods such as deep learning can discover the features of a dataset for themselves, rather than requiring a programmer to define each one. Layered systems can also combine simple features into more abstract ones. The field has a long and uneven history: early neural networks were loosely inspired by biological brains, but current networks are generally regarded as poor models of how brains actually work. Hardware limits and a 1969 book by Marvin Minsky and Seymour Papert, which highlighted the restrictions of single-layer perceptrons, dampened interest during the late 1960s and 1970s.
Machine learning also has a close link with data compression. A system that predicts the probabilities of a sequence from its history can be used to compress that data optimally, and an optimal compressor can in turn be used for prediction. Some researchers have used this equivalence to argue that compression is a sensible benchmark for general intelligence, though this view is not settled. Quantum machine learning, which studies quantum algorithms for learning tasks, is one of several newer branches of the field.
Opens the Intellect+ feed with this card first. Swipe up for the next one.
Sources
- Summary by Intellect · source: Quantum machine learning
- Also: Machine learning in biology - Wikipedia
- Also: Machine learning - Wikipedia
- Also: Outline of machine learning - Wikipedia
- Also: Deep learning
- Also: Neural network (machine learning)
- Also: www.nature.com — Quantum machine learning - Nature
To cite this card, link to https://intellectplus.co.uk/app/technology-machine-learning-2.
How we source, check and correct cards: Editorial standards.
General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.
Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?
Related
More in Researched · More in Technology · Latest · A–Z · All topics