Thurs 19 Jan 7pm at the NYC Machine Learning meetup.
Stan‘s entirely publicly funded and open-source and it has no secrets. Ask us about it and we’ll tell you everything you might want to know.
P.S. And here‘s the talk.
Thurs 19 Jan 7pm at the NYC Machine Learning meetup.
Stan‘s entirely publicly funded and open-source and it has no secrets. Ask us about it and we’ll tell you everything you might want to know.
P.S. And here‘s the talk.
I would like to participate, but I live in Brazil. will be any web based access to the meeting?
Manoel
I confess that I’m a bit confused as to the functionality of this program. Will this program provide the flexibility and functionality of WinBUGS/JAGS etc, but with better convergence properties? Will it replace WinBUGS/JAGS, or does it have limitations that these older programs don’t have?
John:
We’re hoping it will be faster, more general, and more flexible than Bugs etc. We’re using an algorithm that should work better on hard problems, and we’re trying to write the program so that it will be easy to expand by writing functions.
Bob’s a very open guy, so in his talk I’m pretty sure he’ll talk a lot about our struggles, the talk won’t just be a sell-job (which is just as well since we’re giving the software out for free).
I agree with Manoel: It would be great if this talk gets recorded! I am sure it would be a very helpful resource for future Stan users and a popular link and your stan site…
I would add my voice to request that the talk be put up on youtube or vimeo. I think in a previous thread you even had the author of JAGS asking whether STAN renders JAGS (and hence BUGS, in my opinion) obsolete or whether STAN was something different.