Density estimation: Difference between revisions
No edit summary |
No edit summary |
||
| (2 intermediate revisions by the same user not shown) | |||
| Line 1: | Line 1: | ||
In machine learning, density estimation is defined as an [[unsupervised learning]] technique. The purpose | In machine learning, density estimation is defined as an [[unsupervised learning]] technique. The purpose of density estimation is to infer the [[probability density function]] (PDF), from observations of a random variable.<ref name="dartmouth">[https://cs.dartmouth.edu/~wjarosz/publications/dissertation/appendixC.pdf Density estimation]</ref> It learns relations among attributes in the data.<ref name="people.cs.pitt">[http://people.cs.pitt.edu/~milos/courses/cs2750-Spring2012/Lectures/class3.pdf Density estimation]</ref> | ||
== Types == | == Types == | ||
* Parametric density estimation: | * Parametric density estimation: It assumes that the data are from a known family of distributions, such as the normal, lognormal, exponential.<ref>[https://v8doc.sas.com/sashtml/insight/chap38/sect23.htm Parametric Density]</ref> | ||
* Non-parametric density estimation: | * Non-parametric density estimation: | ||
Latest revision as of 19:45, 31 March 2020
In machine learning, density estimation is defined as an unsupervised learning technique. The purpose of density estimation is to infer the probability density function (PDF), from observations of a random variable.[1] It learns relations among attributes in the data.[2]
Types
- Parametric density estimation: It assumes that the data are from a known family of distributions, such as the normal, lognormal, exponential.[3]
- Non-parametric density estimation:
Terminology
- Estimator
- Consistent estimator
- Unbiased estimator
- Parametric methods
- Non-parametric methods
- Explicit density estimation
- Implicit density estimation