Where communities thrive

Join over 1.5M+ people
Join over 100K+ communities
Free without limits
Create your own community

Explore more communities

Rainier/Lobby

People

Repo info

Activity

Jan 31 2019 19:14

avi-stripe on develop

Generated new Bazel BUILD files… (compare)
Jan 31 2019 19:14

avi-stripe closed #322
Jan 30 2019 23:36

areese-stripe review_requested #322
Jan 30 2019 23:36

areese-stripe opened #322
Jan 30 2019 23:33

areese-stripe on areese_bazel_update

Generated new Bazel BUILD files… (compare)
Jan 30 2019 23:33

areese-stripe added as member
Jan 27 2019 02:42

avibryant on distributeconfig

wip (compare)
Jan 23 2019 19:40

avi-stripe on develop

Fix compute bug with batched ob… (compare)
Jan 23 2019 19:40

avi-stripe closed #321
Jan 23 2019 19:34

mio-stripe labeled #321
Jan 22 2019 18:24

mg-stripe added as member
Jan 19 2019 11:26

mg-stripe removed as member
Jan 17 2019 20:02

avi-stripe on develop

stack safe Packer (#273) (compare)
Jan 17 2019 20:02

avi-stripe closed #273
Jan 17 2019 20:01

avi-stripe synchronize #321
Jan 17 2019 20:01

avi-stripe on targettest

fix sbcbenchmark (compare)
Jan 17 2019 19:57

avi-stripe assigned #321
Jan 17 2019 19:57

avi-stripe opened #321
Jan 17 2019 18:58

avi-stripe on targettest

fix bug with batchBits output o… (compare)
Jan 14 2019 22:36

avi-stripe on targettest

failing test for fit gamma targ… (compare)

Avi Bryant

@avibryant

yeah, though again that could be easily reintroduced and probably should be. It was just easier for me to delete with a broad brush and reintroduce as needed. Do you actually use that?

@kailuowang ^

Darren Wilkinson

@darrenjw

I would use that, too!

ie. have "show" use EvilPlot's "displayPlot" method to pop up the plot from the Scala REPL. Does that make sense?

Avi Bryant

@avibryant

oh I see, rather than ascii plots like we had before?

yes that would be great

created stripe/rainier#488, I probably won't work on this right now but PRs very welcome

(the original motivation for the ascii plots was for tut docs, but mdoc makes it easier to support images)

Avi Bryant

@avibryant

BTW do you guys use ammonite or sbt console or what, as a REPL?

Darren Wilkinson

@darrenjw

I typically just use sbt console (old-school, I know)

Avi Bryant

@avibryant

ok I asked this to some others via DM but posting here in case maybe @darrenjw has thoughts:

Screen Shot 2020-02-25 at 11.05.33 AM.png

this is me trying to understand something basic about mass matrix adaptation and HMC

Darren Wilkinson

@darrenjw

I'm not sure I completely understand the problem. But when thinking about this stuff conceptually, I often find it helpful to think about the dynamics of the process in continuous time. In continuous time there is no leap frog, no step size, and there is no step-size adaptation, but there are still good and bad mass matrices, and still good and bad integration times, though these could be related.

Avi Bryant

@avibryant

yes, agreed. But in the discrete world, I guess one way of framing my question is: is there any reason we don't normalize mass matrices (eg, for a diagonal matrix, such that the elements are all <= 1?)

I realize this is a bit hand wavy but it seems like that's beneficial numerically to the leapfrog

but it's clearly not common practice so I fear I'm missing something.

I guess maybe, more correctly, such that the diagonal elements treated as a vector have a length of 1?

Darren Wilkinson

@darrenjw

Or trace 1, would have been my instinct (total variance).

But it feels like it shouldn't be necessary

Avi Bryant

@avibryant

so apart from empirical observations, which could be related to implementation bugs etc, the reason it seems theoretically useful to me is that

(thinking out loud a bit here)

Darren Wilkinson

@darrenjw

It feels like in the case of an MVN target, the mass matrix should reflect the covariance/precision matrix, and not some normalised version. But this isn't something I've implemented in practice, so I could easily be missing something important...

Avi Bryant

@avibryant

in the leap frog, we advance the parameters by stepSize * (momentum * variance)

and then we advance the momentum by stepSize * gradient(parameters)

where the gradient is scaled to a delta of 1 on the parameters

Darren Wilkinson

@darrenjw

Yes it's certainly true that bad choices can make the numerics stiffer, but I feel like the good choices make the numerics better.

Avi Bryant

@avibryant

and I guess the thing that's getting me right now is that if variance is very far from 1,

does that make a good choice of stepSize for the parameter full step,

a bad choice of stepSize for the momentum half step?

I guess I can't really justify why normalizing it would be any better or worse.

Darren Wilkinson

@darrenjw

But my feeling is that you would only choose a variance far from one if different step sizes for the parameters and momentum are appropriate. But I'm seriously hand-waving at this point!

Avi Bryant

@avibryant

ok, I can see the intuition there, I think: if a given parameter has low variance, then its component of the gradient will also be small

and so you want a large stepSize in both cases

Darren Wilkinson

@darrenjw

Yes, I think so.

Avi Bryant

@avibryant

empirically though I'm having a lot of trouble with it. Which could mean there's something wrong with step size adaptation I guess.

Darren Wilkinson

@darrenjw

I would definitely shrink (heavily) towards the identity, and you might additionally want some hard bounds on how big or small masses can get. In high dimensions, estimating covariance matrices stably and well is notoriously difficult, generally, but estimating them from badly tuned MCMC traces is obviously even worse.

Avi Bryant

@avibryant

thanks for the thoughts. I'll keep banging my head on it :)

Darren Wilkinson

@darrenjw

Thanks for your efforts! It's greatly appreciated.

Avi Bryant

@avibryant

for posterity, the problem was this: https://twitter.com/avibryant/status/1238321823839711234

Darren Wilkinson

@darrenjw

:thumbsup:

Zhenhao Li

@Zhen-hao

hi, I haven't tried Rainier yet. but is it possible to sample from a user-defined non-uniform discrete distribution?

for example, with numpy you can do numpy.random.choice(actions, size=len(actions), replace=False, p=my_sampling_probabilities_vector)

Zhenhao Li

@Zhen-hao

where actions are the values of the random variable whose distribution is the my_sampling_probabilities_vector

Avi Bryant

@avibryant

@Zhen-hao there's currently no support for discrete parameters. If you're just looking to generate from a non-uniform discrete distribution, we had support for that in 0.2.x that I removed in 0.3.x in anticipation of adding it back in when I do discrete parameters :/

either way, I'd love to hear what your use case is

Zhenhao Li

@Zhen-hao

@avibryant thanks for your quick response. the use case is that some businesses want specifically defined discrete distributions such as that based on inversed rank (in recommendation system)

for now, it is not hard for me to write one using a uniform distribution U(0,1) for my use case.

so I will go down that path

Avi Bryant

@avibryant

sounds good, let me know if you run into any problems