Bayesian Probability for Babies

BAYESIAN PROBABILITY for babies Chris Ferrie Take a bite. It has no candy. Pr( [no-candy bite] | [no-candy cookie] ) = 1 The probability of a no-candy bite, given a no-candy cookie, is 1. Pr( [no-candy bite] | [candy cookie] ) = 1/3 The probability of a no-candy bite, given a candy cookie, is 1/3. and all had candy but one? 1/3 of the candy cookies have a no-candy bite. This is the posterior distribution of cookies.

A hand flips through the children’s book “Bayesian Probability for Babies,” using illustrated cookies and candy to explain conditional probability. A bite without candy is first judged likely to come from a plain cookie, but after considering that nearly all cookies contain candy, the posterior probability reverses the conclusion: the bite probably came from a candy cookie. The clip humorously turns a sophisticated Bayesian lesson into a simple baby-book demonstration. EN: A hand flips through the children’s book “Bayesian Probability for Babies,” using illustrated cookies and candy to explain conditional probability. A bite without candy is first judged likely to come from a plain cookie, but after considering that nearly all cookies contain candy, the posterior probability reverses the conclusion: the bite probably came from a candy cookie. The clip humorously turns a sophisticated Bayesian lesson into a simple baby-book demonstration.

By using probability for babies This is a ball The ball is a piece of candy on a cookie yum Some cookies have candy some don't Take a bite. It has no candy. Did it come from a candy cookie? Either it came from a candy cookie or it didn't what are the chances? If a cookie had no candy then every bite would have no candy The probability of a no-candy bite given a no-candy cookie is 1. If the cookie had candy then very few bites would have no candy. The probability of a no-candy bite given in a cookie is 1 out of 3. 1 is greater than 1 out of 3. So the no-candy bite probably came from a no-candy cookie. But what if we knew there were 10 cookies and all had candy but one? Take a bite of each There are four no candy bites three bites are from candy cookies and one bite is from a no candy cookie One out of three of the candy cookies have a no candy bite Three out of four the probability of a candy cookie the no candy bite is three out of four This is the prior distribution of cookies. This is the posterior distribution of cookies. This bite probably came from a candy cookie. Now you know by using probability. The end.

• @ithueti / 7133 2022-07-22