Doing Bayesian Data Analysis On the GPU: Examples

[dragandj]

Characteristic examples from the book Doing Bayesian Data Analysis 2nd edition [1] programmed in Clojure and OpenCL to run on the GPU. Much, much faster than Stan or JAGS!

The library used (Bayadera) is still pre-release, so much polishing is still needed, so this can be considered a preview. But, it is still very useful, and not more complex for programmers than the mainstream Bayesian tools.

[1] https://www.amazon.com/Doing-Bayesian-Data-Analysis-Second/d...

[feral]

Any hint how much faster? Not looking for defensible benchmarks, but are you talking an order of magnitude? Multiple orders?

Does it make the same probabilistic guarantees as the methods used in Stan etc? Or is it trading validity for speed?

[marmaduke]

It's not really comparable to stan if you're not computing gradients for HMC. Of course maybe for these models you don't need to.

[te]

Looks cool. Would love to see the "much, much faster" claim quantified. Both including and excluding compile times. Stan is neat, but the recompile time after every tweak to the model really got to be a drag. If you can improve on that, it would be a real win for me.

[thom]

This looks great, and very tastefully implemented on the Clojure side. Have you considered splitting out things like the quil utils?

[anon1253]

Just of out curiosity, how hard would it be to write a compiler that takes JAGS or STAN model files and compiles it to the s-exps needed to use the library?

[jsweojtj]

I must be missing something. Is there any output that we are supposed to be seeing? Plots of the fits or something?

[dragandj]

Yes. Run the tests in the REPL. Call (analysis) to only invoke the fits and get the timings, or call (display-sketch) and then (reset! all-data (analysis)) to see the plots. It is only done like that in these examples; for actual work, you are free to use any meaning combination of functions that you like.