课程: Probability Foundations for Data Science

今天就学习课程吧！

今天就开通帐号，24,700 门业界名师课程任您挑！

Maximum a posteriori estimation (MAP)

Maximum a posteriori estimation (MAP)

课程: Probability Foundations for Data Science

开始一个月试用 购买团队方案

Maximum a posteriori estimation (MAP)

“

- [Instructor] Let's move on to the next estimation technique. Maximum A Posteriori Estimation, also known as MAP, is a method used in Bayesian probability to estimate an unknown parameter by maximizing the posterior distribution. This parameter equals the mode of the posterior distribution. It combines the prior information about the parameter with the likelihood of the observed data to provide a more robust estimation. Maximum A Posteriori Estimation is closely related to Maximum Likelihood Estimation. The Maximum A Posteriori Estimation method employs an augmented optimization objective that incorporates a prior distribution to quantify additional information available through prior information about the parameter being estimated. Maximum A Posteriori Estimation is therefore a regularization of Maximum Likelihood Estimation. This is due to the prior information being incorporated. Related to your earlier work with Bayes' Theorem, the Posterior Distribution combines the prior…

内容