The limits of MAP inference by MWSS on perfect graphs
Add to your list(s)
Download to your calendar using vCal
If you have a question about this talk, please contact dsu21.
I shall briefly describe a recent, promising method to perform MAP inference for discrete undirected graphical models based on reducing the optimization problem to finding a maximum weight stable set (MWSS) in a derived weighted graph, which if perfect, may be performed in time polynomial in the number of variables. I shall discuss recent work (AISTATS 2015), where the limits of this approach were established for the class of binary pairwise (Ising) models, yielding a simple, interesting characterization.
http://jmlr.org/proceedings/papers/v38/weller15.pdf
This talk is part of the Machine Learning @ CUED series.
This talk is included in these lists:
Note that ex-directory lists are not shown.
|