You are not logged in.
I came accross this interactive lesson on Bayesian inference on Github. I'm reading through it now and I find it to be quite good. I've known the basics of Bayesian statistics for a while but I've never had to use it in my research so I've never looked into how to actually implement it. It looks like this book is an excellent overall introduction, with a stronger focus on programming (Python) than on the underlying mathematics.
Anyway, I thought I'd share it since it might be interesting for some of you:
https://github.com/CamDavidsonPilon/Pro … or-Hackers
Last edited by jakobcreutzfeldt (2013-06-06 12:56:57)
Offline
I personally haven't seen much programming based on propositional logic at all. Thanks for sharing this, as it is certainly something that I could benefit from learning more about.
Offline
Thanks for posting this! I know rather little about the topic, and this looks like a good introduction. I'll probably want more mathematical rigour before long, but as the author notes, one can always move on to more formal texts later.
Officer, I had to drive home - I was way too drunk to teleport!
Offline
http://www.paulgraham.com/better.html
Also Allen Downey has a great book called Think Stats: http://www.greenteapress.com/thinkstats/
and another one in the works called Think Bayes: http://www.greenteapress.com/thinkbayes/
The first one will also teach you about statistical hypothesis testing in the frequentist tradition (ala R.A. Fisher)
Last edited by Nisstyre56 (2013-06-16 23:44:11)
In Zen they say: If something is boring after two minutes, try it for four. If still boring, try it for eight, sixteen, thirty-two, and so on. Eventually one discovers that it's not boring at all but very interesting.
~ John Cage
Offline
I know a bit about probability. But could you dumb this down a notch for me? What can it actually be used for? Machine learning in particular maybe, but what else?
Offline
I know a bit about probability. But could you dumb this down a notch for me? What can it actually be used for? Machine learning in particular maybe, but what else?
Scientists make use of it when designing experiments and you can make use of it to simulate said experiments. You can also use probability theory to create models of stochastic (somewhat random) phenomena like English text, and then use the "parameters" (numbers) of that model to generate something statistically similar to English. This is called a Markov Model. I suggest giving some of the stuff on Peter Norvig's site a read: http://norvig.com
In Zen they say: If something is boring after two minutes, try it for four. If still boring, try it for eight, sixteen, thirty-two, and so on. Eventually one discovers that it's not boring at all but very interesting.
~ John Cage
Offline