[ / / / / / / / / / / / / / ] [ dir / random / 93 / biohzrd / hkacade / hkpnd / tct / utd / uy / yebalnia ]

/templeos/ - The 64-Bit Temple Operating System

Discuss God and Code.

Name
Email
Subject
REC
STOP
Comment *
File
Password (Randomized for file and post deletion; you may also set your own.)
Archive
* = required field[▶Show post options & limits]
Confused? See the FAQ.
Flag
Embed
(replaces files and can be used instead)
Options
dicesidesmodifier

Allowed file types:jpg, jpeg, gif, png, webp,webm, mp4, mov, swf, pdf
Max filesize is16 MB.
Max image dimensions are15000 x15000.
You may upload5 per post.


File: 1427098799994.png (277.72 KB,1090x819,1090:819,BayesTheoremSmallerPosterZ….png)

 No.1319

Hello Terry. Thanks for building the Temple. Do you know Bayes' Theorem?

If so, what do you think of it? Is it a valid way to deal with the Random?

If not: http://oscarbonilla.com/2009/05/visualizing-bayes-theorem/

And, if that dude’s diagrams don’t go down easy, here’s how I’d put it:

Any potential event you want to know about, given that you’ve found a new piece of evidence for it happening, is exactly as likely as: however likely it is the evidence happens whenever the event happens, times however likely it is the event happens at all, divided by how likely it is that the evidence happens at all.

Since all evidence comes from events and all events can be counted as evidence, you can switch the words and it means the same thing.

This incantation works even when you are not certain about the exact probabilities — even with estimates, it will still tell you in which direction to move your thoughts. And of course, how confident you are in those estimates depends on how you review the evidence for them…

You can even start with totally random percentages and revise as you find new evidence. Unlike the traditional hypothesis-experiment-conclusion loop as taught in kiddy science, this lets you figure out how uncertain you should be. About anything.

Other ways to say it:

probability of A given B = ( (prob of B given A) * (prob of A) ) / (prob of B)

I saw evidence of an event. Did the event really happen? Well:

-If the the evidence is common whenever the event happens (prob of B given A), that raises the likelihood that your evidence means the event happened.

-If the event is common even without the evidence (prob of A), that raises the likelihood that your evidence means the event happened.

-If the evidence is common even without the event (prob of B), that lowers the likelihood that your evidence means the event happened.

*All three of these must be taken into account.*

-You can swap ‘event’ with ‘evidence’ and it means the same thing. You can swap ‘raises’ and ‘lowers’ if you replace ‘common’ with ‘rare.’

-An event not happening is just another event, if it helps to think in reverse.
____________________________
Disclaimer: this post and the subject matter and contents thereof - text, media, or otherwise - do not necessarily reflect the views of the 8kun administration.


[Return][Go to top][Catalog][Nerve Center][Random][Post a Reply]
Delete Post [ ]
[]
[ / / / / / / / / / / / / / ] [ dir / random / 93 / biohzrd / hkacade / hkpnd / tct / utd / uy / yebalnia ]