News for hackers. Stuff that matters.

A comment on Bayesian statistics and machine learning: How do they differ?. Parent comment

Re:Bayesian statistics and machine learning: How do they dif... by lalaland1125 on Sunday January 15, 2023 @10:38AM (#34390144)
A "posterior probability for p" for a Bernoulli distribution is meaningless because it's equivalent to a single Bernoulli distribution.
[ Parent | on HN ]
Re:Bayesian statistics and machine learning: How do they dif... by steppi on Sunday January 15, 2023 @10:42AM (#34390175)
If one only had such an estimate for a single example, this would be true, but in aggregate over many predictions, the uncertainty bands can useful for decision making. This is an active area of research.
[ Parent | on HN ]
Re:Bayesian statistics and machine learning: How do they dif... by contravariant on Sunday January 15, 2023 @10:55AM (#34390285)
It's meaningful in the sense that what the model produces is often not the posterior probability.

See my other comment for more detail.

[ Parent | on HN ]