AIRG,
I would like to remind you that this afternoon David Merrell will be
giving an overview of variational inference, a core category of
approximate inference. Variational inference is often more efficient and
more scalable than MCMC, making it essential for tackling large
problems. This introduction from two of the biggest names in the field
can only serve to increase your knowledge and understanding of this
complex (and therefore often underappreciated) topic.
4pm, CS *3310*
https://doi.org/10.1080/01621459.2017.1285773
http://www.cs.columbia.edu/~blei/fogm/2018F/materials/BleiKucukelbirMcAuliffe2017.pdf
Aubrey
On 3/25/19 9:44 AM, DAVID MERRELL wrote:
> Hi AIRG,
>
> In this week's AIRG, I'll lead a discussion about this 2017 paper by
> David Blei et al:
>
> "*Variational Inference: A Review for Statisticians*"
> https://doi.org/10.1080/01621459.2017.1285773
>
> You don't need to be a statistician to enjoy this paper!
> Variational inference reformulates Bayesian inference as an optimization
> problem.
> It's gained widespread use in the world of Bayesian computing and
> probabilistic programming.
> It's definitely worth your while to know something about this!
>
> See you there at *4:00pm Wednesday, 4310 CS.*
> David Merrell
>
> _______________________________________________
> AIRG mailing list
> AIRG@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/airg
>
|