machine learning
In plain wordsA way of building software that learns from examples instead of following rules written by a programmer. Shown thousands of labeled photos, for instance, a program can learn to recognize cats.
The branch of artificial intelligenceartificial intelligence, n.: Computer systems that do things we would normally credit to a thinking mind, such as holding a conversation, reading an X-ray or planning a route, and the field of research that builds them. in which a program's behavior is derived from data rather than written out by hand: an algorithmalgorithm, n.: A precise, step-by-step procedure for solving a problem or doing a task, like a recipe a computer can follow. In the news, 'the algorithm' often means software that ranks what you see online. adjusts the parametersparameter, n.: One of the adjustable numbers inside an AI model that are tuned during training. Large language models have billions of them, and together they store what the model has learned. of a modelmodel, n.: A simplified working version of something. In AI it usually means the trained program, a big set of learned numbers, that takes an input and produces an answer, prediction or piece of content. until it fits training datatraining data, n.: The examples an AI system learns from. Whatever patterns, skills, gaps and biases are in the training data tend to show up in the finished system., and the fitted model then makes predictionsprediction, n.: A model's output for a given input: its best guess at an unknown answer. In AI it need not be about the future; labeling a photo 'cat' counts as a prediction. or decisions about inputs it has never seen. Tom Mitchell's widely cited formulation names three ingredients, a class of taskstask (machine learning), n.: The job an AI model is meant to do, such as translating sentences, spotting spam or answering questions, defined by its inputs, its expected outputs and how success is scored., a performance measure and experience, and counts a program as learning when its measured performance on those tasks improves with experience. The main settings are supervised learningsupervised learning, n.: Teaching a computer by example: you give it many inputs paired with the right answers, such as photos labeled 'cat' or 'dog', and it learns to answer correctly for new inputs. (labeled examples), unsupervised learningunsupervised learning, n.: Machine learning without answer keys: the computer looks at unlabeled data and finds structure on its own, such as groups of similar customers or the main ways in which data vary. (structure in unlabeled data), self-supervised learningself-supervised learning, n.: A way for AI to learn from huge amounts of unlabeled data by setting itself fill-in-the-blank puzzles, such as guessing a hidden word or a missing patch of an image. and reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time. (rewards from an environment). Methods range from linear regressionlinear regression, n.: Fitting a straight line (or flat surface) through data points to predict one number from others — say, a house's price from its floor area — choosing the line that keeps prediction errors smallest. and decision treesdecision tree, n.: A model that makes a prediction by asking a series of yes-or-no questions about the input, like a flowchart, with each answer leading to the next question until it reaches a final answer. to neural networksneural network, n.: A kind of computer program that learns a task from examples by tuning thousands to billions of connections between simple number-crunching units, an arrangement loosely inspired by nerve cells in the brain.; deep learningdeep learning, n.: Machine learning done with neural networks built from many stacked layers: raw data such as pixels, sound or text goes in at one end, and each layer turns what it receives into something more abstract. is the subfield that uses many-layered networks. The central difficulty is generalizationgeneralization, n.: How well a model copes with new examples it has never seen, rather than just memorizing its training data, like a student who understands the method, not just the answers.: performing well on data not seen in training rather than memorizing it (overfittingoverfitting, n.: When a model learns its training examples too well — quirks and noise included — so it does great on data it has seen but poorly on anything new, like a student who memorizes answers without understanding.).
We replaced the hand-tuned fraud rules with a machine learning model trained on five years of transactions.
Origin
Usually credited to Arthur Samuel of IBM, who used the phrase in the title of his 1959 paper 'Some Studies in Machine Learning Using the Game of Checkers', describing a checkers program that improved with experience, partly by playing thousands of games against itself. Machine learning consolidated as a discipline from 1980, with the first machine-learning workshop at Carnegie Mellon (counted as the first ICML), the journal Machine Learning (1986) and the NIPS conference (from 1987; now NeurIPS).
Usage note
The often-quoted definition 'the ability to learn without being explicitly programmed' is attributed to Samuel but does not appear in his 1959 paper. Statisticians call much of the same territory statistical learning. Machine learning is a subset of artificial intelligenceartificial intelligence, n.: Computer systems that do things we would normally credit to a thinking mind, such as holding a conversation, reading an X-ray or planning a route, and the field of research that builds them., though since the 2010s the two are often used interchangeably.
Related
- artificial intelligenceartificial intelligence, n.: Computer systems that do things we would normally credit to a thinking mind, such as holding a conversation, reading an X-ray or planning a route, and the field of research that builds them.
- deep learningdeep learning, n.: Machine learning done with neural networks built from many stacked layers: raw data such as pixels, sound or text goes in at one end, and each layer turns what it receives into something more abstract.
- supervised learningsupervised learning, n.: Teaching a computer by example: you give it many inputs paired with the right answers, such as photos labeled 'cat' or 'dog', and it learns to answer correctly for new inputs.
- unsupervised learningunsupervised learning, n.: Machine learning without answer keys: the computer looks at unlabeled data and finds structure on its own, such as groups of similar customers or the main ways in which data vary.
- reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time.
- trainingtraining, n.: The process by which an AI model learns: it is shown huge numbers of examples and its internal settings are adjusted, bit by bit, until its answers get better.
- generalizationgeneralization, n.: How well a model copes with new examples it has never seen, rather than just memorizing its training data, like a student who understands the method, not just the answers.
- data sciencedata science, n.: The practice of collecting, cleaning, analyzing and interpreting data, using statistics, programming and machine learning, to answer questions and guide decisions.
- symbolic AIsymbolic AI, n.: An approach to AI that writes knowledge down as human-readable symbols and rules, such as 'every bird has wings', and reasons by manipulating them step by step instead of learning patterns from data.
- rule-based systemrule-based system, n.: A computer system that makes decisions by following a list of 'if… then…' rules written by people, rather than by learning from examples.
Sources
- Samuel, A. L. (1959). Some Studies in Machine Learning Using the Game of Checkers. IBM Journal of Research and Development, 3(3), 210–229.
- Mitchell, T. M. (1997). Machine Learning. McGraw-Hill.
Defined in law or standards
US Executive Order 14110 (2023; revoked January 2025), §3(t)
Mentioned in
AI winterAI winter, n.: A slump in which disappointment over AI's unmet promises leads governments and companies to pull back money and attention. Historians count two big ones: one in the 1970s and one starting in the late 1980s., artificial intelligenceartificial intelligence, n.: Computer systems that do things we would normally credit to a thinking mind, such as holding a conversation, reading an X-ray or planning a route, and the field of research that builds them., datadata, n.: Recorded facts, measurements, text, images or sounds that a computer can store and process. AI systems learn their skills from large amounts of data rather than from rules written by hand., deep learningdeep learning, n.: Machine learning done with neural networks built from many stacked layers: raw data such as pixels, sound or text goes in at one end, and each layer turns what it receives into something more abstract., reinforcement learningreinforcement learning, n.: A way of teaching a computer program by trial and error: it tries actions, earns points for good results and penalties for bad ones, and gradually learns which choices pay off over time., roboticsrobotics, n.: The science and engineering of building robots and making them work: designing their bodies, sensors and motors, and writing the software that lets them see, plan and move safely., AI literacyAI literacy, n.: Knowing enough about AI to use it sensibly and judge it critically: what it can and cannot do, how it goes wrong, and how it affects people., AI-poweredAI-powered, adj.: Described as using artificial intelligence to work. The label appears on everything from search engines to toothbrushes and often says little about how much AI is really involved., algorithmalgorithm, n.: A precise, step-by-step procedure for solving a problem or doing a task, like a recipe a computer can follow. In the news, 'the algorithm' often means software that ranks what you see online., Andrew NgAndrew Ng, prop. n.: A British-born American computer scientist whose online courses have taught machine learning to millions; he also co-founded Google's deep learning team, Google Brain, and the online-course company Coursera., Arthur SamuelArthur Samuel, prop. n.: An American IBM engineer whose checkers program improved by playing against itself, and whose 1959 paper helped make 'machine learning' the name of a field., Computing Machinery and IntelligenceComputing Machinery and Intelligence, prop. n.: Alan Turing's 1950 essay asking whether machines can think. It proposed a conversation test in place of that question, answered objections to thinking machines and suggested that machines might learn like children., DARPA Grand ChallengeDARPA Grand Challenge, prop. n.: A series of US government-sponsored races for driverless vehicles held in 2004, 2005 and 2007. No robot finished the first race; within three years, cars were navigating mock city traffic., data miningdata mining, n.: Searching large collections of data with computers to discover useful patterns, such as products often bought together or signs that a transaction may be fraudulent., data sciencedata science, n.: The practice of collecting, cleaning, analyzing and interpreting data, using statistics, programming and machine learning, to answer questions and guide decisions., expert systemexpert system, n.: A program that stores a human specialist's know-how as many if–then rules and uses them to give advice or make decisions in a narrow field, such as diagnosing infections or configuring computers., good old-fashioned AIgood old-fashioned AI, n.: A nickname for the classic style of AI, dominant from the 1950s to the 1980s, that tried to make computers intelligent by writing knowledge and rules down by hand., IBM WatsonIBM Watson, prop. n.: An IBM question-answering computer that beat two champions of the US quiz show Jeopardy! in 2011. IBM then sold business AI products under the Watson name, many of which disappointed., ICMLICML, prop. n.: A leading machine learning conference, first held as a workshop in 1980, where researchers gather each summer to present new learning methods and theory., inductive reasoninginductive reasoning, n.: Reasoning from particular examples to a general rule or prediction — every swan I have seen is white, so swans are probably white — which is useful but can always turn out wrong., learninglearning, n.: Getting better at something through experience. For a machine, it means changing its internal settings based on data so that it performs a task better than before., Leo BreimanLeo Breiman, prop. n.: An American statistician who invented random forests and bagging, decision-tree methods still in wide use, and urged statisticians to judge models by how well they predict., machine learning engineermachine learning engineer, n.: An engineer who turns machine-learning ideas into working systems: preparing data, training and testing models, and running them reliably inside real products., MLOpsMLOps, n.: The working habits and tools that keep machine-learning models dependable inside real products once the experiment is over: automated testing, release, monitoring and retraining; the factory and maintenance crew around a model, not the model itself.