Yay Stan! So what's the deal with BBVI? I heard it's integrated in the newest Stan, but can I use it? What for?
396 karma · joined February 21, 2009
The second is to go out and meet the practitioners where they are. That means making Stan models look like the equations in your paper. It means connecting to existing research software ecosystems. It means making it easy to adapt. In this case, if you end up tightly coupled to, say, specific R data structure then it becomes much, much harder to make it also work well for Python, and for Stata, etc. Additionally, it means that if I start on a model in Python then I cannot easily switch to using R (or vice-versa).
Using a DSL also has the nice property of cleanly separating Stan world vs R world. That makes it easier for users to build an accurate mental model for what happens when/where, which hopefully then leads to better performance.
Most importantly though, Stan code is incredibly concise and clear for specifying models.
For those who haven't used it, the typical usage is to use Stan in combination with other languages (most commonly R). In the Stan language you define a model, specifying the data you'll get, potentially transformations to that data, then a set of parameters you want to fit, then finally a model saying how those parameters interact with each other and the data. You can also define priors for parameters. Then typically you save that model, and using R pass in the variables to a Stan call. The resulting fit object is returned to your original environment automatically.
It actually fits the work style reasonably well. All data munging happens as before, but instead of having some complicated model expression in, say, a glm() call, you have it in Stan language.
If you're interested in this, Gelman has two great books. Gelman & Hill's Hierarchical Models[2] which is applied and geared towards social science researchers, and Bayesian Data Analysis[3].
[1] https://www.kickstarter.com/projects/fred/the-emoji-translat...
As an Ars forum member from back as early as 2001, I appreciate everything you did, John. Thanks!
Massive amounts of risk, supply chain, logistics, all eliminated when selling to the UN like this.
Overall I'm a fan of the Dreamliner (only 2 flights, a roundtrip on Norwegian). Seats were reasonably comfortable, even for economy on a budget airline. The big windows are nice. It was definitely quieter and (placebo?) the air did feel less gross.
Downsides: The arm rest didn't fully rise out of the way, so it's no good for sleeping on your partner's shoulder. The not-fully-dimming windows were a little annoying, but also really neat.
https://www.youtube.com/watch?v=ZwwS4YOTbbw
https://github.com/search?utf8=%E2%9C%93&q=__init__&type=Cod...
Crazier, IMO, is the DAU. 829 million log in EVERY DAY.
He's on a quixotic quest toward a certain kind of world, with openness and transparency. Unfortunately when he tries to interface with the rest of the world it can often go horribly wrong. His general attempts at embracing those who disagree with him, eg by thanking them and retweeting, exacerbate and lend weight to unpleasant comments.
It's truly unfortunate that Chad didn't manage to support the diversity activist community in a way that was acceptable to them. I think fundamentally there missions are much more similar that they're acknowledging now, and each is loosing out on what could potentially be a great ally.
I hope in the future these sort of things can be more constructively handled on both sides that either has so far.