prediction
In plain wordsA 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.
The output a trained model produces for an input: a class label, a probability, a number, a ranking or, for generative modelsgenerative model, n.: A model that learns what a kind of data looks like — faces, sentences, songs — well enough to produce new examples that resemble the originals without copying them., the next tokentoken, 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. or a denoised image. In machine learningmachine learning, n.: A 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. 'prediction' covers any estimate of an unobserved quantity, not only a future one; predicting whether a scan shows a tumor concerns the present. Predictions may be point estimates or full probability distributions, and their quality is judged by metrics such as accuracyaccuracy, n.: The share of a model's answers that are right, such as 95 out of 100. Simple and intuitive, but misleading when one kind of answer is far more common than the others. or mean squared errormean squared error, n.: A way of scoring numerical predictions: square each prediction's error, then average them, so big misses count much more than small ones. Lower is better. and by calibrationcalibration, n.: How well a model's stated confidence matches reality: when a well-calibrated system says it is 80% sure, it turns out to be right about 80% of the time. (whether stated probabilities match observed frequencies). Statisticians distinguish prediction (accurate outputs for new cases) from explanation and causal inferencecausal inference, n.: Working out whether one thing actually causes another, rather than merely occurring alongside it — for example, whether a new feature made users stay longer or just happened at the same time. (understanding why): a model can predict well by exploiting spurious correlations, so good predictions do not by themselves justify interventions.
The model's prediction was 'benign' at 0.62 confidence, low enough to trigger a second read.
Origin
From Latin praedicere, 'to say beforehand'. Statistics long used 'prediction' for estimating unobserved values, and machine learning adopted it for model outputs generally. Leo Breiman's 2001 essay 'Statistical Modeling: The Two Cultures' contrasted the prediction-focused culture of machine learning with traditional statistical data modeling.
Usage note
A prediction becomes a decision only when a threshold or policy acts on it. Calling language modelslarge 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. 'next-token predictors' names their training objective; what that implies about their abilities is debated.
Related
- inferenceinference, n.: Drawing a conclusion from evidence. In today's AI it most often means running an already-trained model to get an answer, as opposed to training it.
- classificationclassification, n.: Sorting things into categories — spam or not spam, which animal is in a photo, which language a text is written in — by learning from examples that are already labeled.
- regressionregression, n.: Predicting a number rather than a category — a house's price, tomorrow's temperature, a patient's blood pressure — by learning from past examples where the answer is known.
- calibrationcalibration, n.: How well a model's stated confidence matches reality: when a well-calibrated system says it is 80% sure, it turns out to be right about 80% of the time.
- predictive AIpredictive AI, n.: AI used to forecast outcomes about people or events, such as who will repay a loan, which patient may get sicker or which machine will fail, in order to guide decisions.
- next-token predictionnext-token prediction, n.: The simple task most AI chatbots were trained on: given some text, guess the next small piece of it. Repeating that guess, piece by piece, produces whole answers.
- time series forecastingtime series forecasting, n.: Predicting future values of something measured over time, such as sales, electricity demand, prices or rainfall, from its past values and related information.
- causal inferencecausal inference, n.: Working out whether one thing actually causes another, rather than merely occurring alongside it — for example, whether a new feature made users stay longer or just happened at the same time.
Sources
- Breiman, L. (2001). Statistical Modeling: The Two Cultures. Statistical Science, 16(3), 199–231.
Mentioned in
machine learningmachine learning, n.: A 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., AI modelAI model, n.: The trained core of an AI: a large file of learned numbers plus the recipe for using them, which turns an input such as a question or a photo into an output., ground truthground truth, n.: The reference answers treated as correct when checking or training an AI system — for example, the confirmed diagnoses that a medical-imaging model's guesses are compared against., inferenceinference, n.: Drawing a conclusion from evidence. In today's AI it most often means running an already-trained model to get an answer, as opposed to training it.