example
In plain wordsA single item of data used to train or test an AI model, such as one photo with its label, one email marked spam or not, or one question with its answer.
Machine learning
A single unit of data in a datasetdataset, n.: An organized collection of examples gathered for a purpose — such as photos with their captions, or questions with answers — used to teach or test an AI system.: an input, such as a feature vector, image, document or conversation, optionally paired with a target labellabel, n.: The correct answer attached to an example in a dataset — 'cat' on a photo of a cat, 'positive' on an upbeat review — that a learning system is trained to predict. or desired output. Training examples are used to fit a model; validation and test examples to tune and evaluate it. Statisticians speak of observations, database users of records or rows, and vision and text researchers of instances. In language-modellarge language model, n.: A computer program trained on enormous amounts of text to predict the words that come next. At vast scale, that simple skill lets it chat, answer questions, translate, summarize and write code. pretrainingpretraining, n.: The first and usually the most expensive stage of building a modern AI model, in which it learns general patterns from enormous amounts of raw text, images or code before being specialized for particular uses. the unit is less clear-cut, since text is cut into fixed-length sequences of tokenstoken, n.: The small chunk of text an AI language model reads and writes — often a word, part of a word or a punctuation mark. AI usage, limits and prices are counted in tokens.; examples given inside a prompt are called shots (few-shot promptingfew-shot prompting, n.: Showing an AI a handful of examples of a task — inputs with the answers you want — inside the prompt, so it can copy the pattern for a new case.). Special kinds include adversarial examplesadversarial example, n.: An input altered on purpose, often too slightly for people to notice, so that an AI model gets it badly wrong — like a panda photo that a classifier confidently calls a gibbon., crafted to cause errors, and hard negatives, near-miss examples used to sharpen a model's distinctions.
Each example is a customer email paired with the department that should handle it.
Origin
Learning 'from examples' is an old theme in psychology and AI, as in Patrick Winston's 1970 MIT thesis on learning structural descriptions from examples. 'Training example' was standard in machine-learning texts by the 1980s.
Usage note
'Sample' is ambiguous: in statistics a sample is a set of observations, in machine learning often a single example, and in generative AI a single generated output, as in 'sampling from the model'. Duplicated examples distort both training and evaluation (data contaminationdata contamination, n.: When the test questions used to grade an AI system were already in its training data, so a high score may reflect memorized answers rather than real ability.).
Related
- datasetdataset, n.: An organized collection of examples gathered for a purpose — such as photos with their captions, or questions with answers — used to teach or test an AI system.
- labellabel, n.: The correct answer attached to an example in a dataset — 'cat' on a photo of a cat, 'positive' on an upbeat review — that a learning system is trained to predict.
- training settraining set, n.: The examples a machine-learning model learns from, like the worked problems a student practices on, kept separate from the examples later used to check how well it learned.
- test settest set, n.: A set of examples kept completely apart from training and used at the end to measure how well a model works on data it has never seen, like a sealed final exam.
- few-shot promptingfew-shot prompting, n.: Showing an AI a handful of examples of a task — inputs with the answers you want — inside the prompt, so it can copy the pattern for a new case.
- adversarial exampleadversarial example, n.: An input altered on purpose, often too slightly for people to notice, so that an AI model gets it badly wrong — like a panda photo that a classifier confidently calls a gibbon.
- data contaminationdata contamination, n.: When the test questions used to grade an AI system were already in its training data, so a high score may reflect memorized answers rather than real ability.
- samplingsampling, n.: Picking some items at random from a larger set or from a range of possibilities — surveying 1,000 voters, say, or letting an AI choose its next word by a weighted lottery.
- featurefeature, n.: One piece of information about each example that a model uses, like a house's size or a pixel's color, or a pattern a model has learned to detect inside itself.
Mentioned in
featurefeature, n.: One piece of information about each example that a model uses, like a house's size or a pixel's color, or a pattern a model has learned to detect inside itself.