## Baidu Encyclopedia version

In probability and statistics theory, the generation model refers to a model that can randomly generate observation data, especially given the conditions of certain implicit parameters. It assigns a joint probability distribution to the observations and the annotated data sequence. In machine learning, the generation model can be used to model data directly (for example, to sample data based on a probability density function of a variable), or to establish a conditional probability distribution between variables. The conditional probability distribution can be formed by the generated model according to Bayes' theorem.

## Wikipedia version

In statistical classification, including machine learning, the two main methods are called**generate**Methods and**Discriminate**method. These computational classifiers use different methods and the degree of statistical modeling is different. The terms are inconsistent, but can be distinguished by three main types, following Jebara (2004):

- Given an observable variable
*X*And target variables*ÿ*,One**Generating model**Is a joint probability distribution of a statistical model*X*×*ÿ*.

- A
**Discriminant model**Yes, the target of the conditional probability of the model*ÿ*Given the observation*X*Symbolically

- Classifiers that do not use probability model calculations are also broadly referred to as "discriminability."

The difference between the last two classes is not consistent; Jebara (2004) calls these three categories*Generate learning*.*Conditional learning*和*Discriminatory learning*, but Ng&Jordan (2002) only distinguishes between two categories, called**Generating classifier**(joint distribution) and**Discriminant classifier**(conditional distribution or no allocation), there is no distinction between the latter two categories. Similarly, based on the generated model**Classifier**Be**Generating classifier**And the classifier based on the discriminant model is**Discriminant classifier**, although the term also refers to a model-based classifier. Each standard example is a linear classifier, which is: Generate Classifier: Naive Bayes Classifier and Linear Discriminant Analysis; Discriminant Model: Logistic Regression; Non-Model Classifier: Perceptron and Support Vector Machine.

## Extended reading

## Comments