model.py, which recomputes what the entry states: the script block by block · open in ColabDictionary of Applied Machine Learning: 114 machine-learning terms, each with a definition, a figure, citations and a Python demo · All terms · Demos · Start here
Updated on 2026-10-29
Readiness 100/100
See also hypothesis space hypothesis training
In machine learning (ML), the word model is used in different ways. Formally, a model is a hypothesis space $\mathcal{H}$: the set of candidate hypothesis maps from which an ML method picks (or learns) one. In applied ML literature and in software libraries, the word model also refers to the trained predictor, i.e., to the learned hypothesis returned by an ML algorithm; the EU AI Act uses this sense in the compound term general-purpose AI model (GPAI model). A different type of mathematical model is a probabilistic model, which specifies a family of probability distributions that characterize a data generation process.
model.py, which recomputes what the entry states: the script block by block · open in ColabP-hypospaceConsider the problem of predicting tomorrow's maximum daytime temperature from today's morning temperature. Machine learning (ML) methods learn a hypothesis map $\widehat{h}$ that maps today's morning temperature, the feature, to an accurate prediction of tomorrow's maximum temperature, the label. This learned hypothesis is chosen from a (typically large) set of candidate hypothesis maps, i.e., a hypothesis space $\mathcal{H}$.
The hypothesis space underlying an ML method is its model: the set of candidate hypothesis maps, chosen in advance, from which the method picks one (Shalev-Shwartz and Ben-David, 2014, Sect. 2.3). For example, linear regression and logistic regression use the linear model as their hypothesis space. A decision tree method uses a hypothesis space that consists of hypothesis maps generated by a flow chart. Large language model (LLM) systems use a hypothesis space consisting of non-linear maps represented by an artificial neural network (ANN).
P-trainedModel training refers to the process of finding a hypothesis in the hypothesis space that yields accurate predictions. A trained (or fitted) model is a learned hypothesis $\widehat{h} = \mathcal{A}(\mathcal{D}^{(\rm train)})$, the output of the map $\mathcal{A}$ by which the ML method turns a training set $\mathcal{D}^{(\rm train)}$ into a hypothesis. In applied ML literature and in software libraries, the word model typically refers to a trained model, i.e., to a specific hypothesis map $\widehat{h}$ (Goodfellow et al., 2016, Sect. 5.1; Murphy, 2012, Sect. 3.5.2; Prince, 2023). The EU AI Act uses the word model in the same sense (i.e., a learned hypothesis) in the compound term general-purpose AI model (GPAI model) (European Parliament and Council of the European Union, 2024, Art. 3(63)).
P-probmodelA different type of mathematical model is a probabilistic model, which specifies a family of probability distributions. Each such probability distribution defines a data generator: data points are realizations of independent and identically distributed (i.i.d.) random variables (RVs) with the given probability distribution.
Fig. 1 illustrates these three uses of
the word model: the hypothesis space, the trained
model, and the probabilistic model.
Synonyms: hypothesis space, hypothesis class, model class.
See also: hypothesis space, hypothesis, training, probabilistic model, parametric model, model parameter, general-purpose AI model.
Published version: doi.org/10.1007/978-981-92-2012-0_2-1, in Dictionary of Applied Machine Learning (Springer Nature Singapore). Please cite that version. This page corresponds to version 1 of that entry.
@incollection{dictml_model,
author = {Jung, Alexander},
editor = {Olioumtsevits, Konstantina and Schnoor, Ekkehard},
title = {model},
booktitle = {Dictionary of Applied Machine Learning},
publisher = {Springer Nature Singapore},
year = {2026},
doi = {10.1007/978-981-92-2012-0_2-1},
isbn = {978-981-92-2012-0},
url = {https://doi.org/10.1007/978-981-92-2012-0_2-1}
}