Timezone: »

 
Poster
Information Theoretic Properties of Markov Random Fields, and their Algorithmic Applications
Linus Hamilton · Frederic Koehler · Ankur Moitra

Wed Dec 06 06:30 PM -- 10:30 PM (PST) @ Pacific Ballroom #41
in Posters Wed »
Markov random fields are a popular model for high-dimensional probability distributions. Over the years, many mathematical, statistical and algorithmic problems on them have been studied. Until recently, the only known algorithms for provably learning them relied on exhaustive search, correlation decay or various incoherence assumptions. Bresler gave an algorithm for learning general Ising models on bounded degree graphs. His approach was based on a structural result about mutual information in Ising models. Here we take a more conceptual approach to proving lower bounds on the mutual information. Our proof generalizes well beyond Ising models, to arbitrary Markov random fields with higher order interactions. As an application, we obtain algorithms for learning Markov random fields on bounded degree graphs on $n$ nodes with $r$-order interactions in $n^r$ time and $\log n$ sample complexity. Our algorithms also extend to various partial observation models.
Keywords: Information Theory · Graphical Models · Unsupervised Learning · Computational Complexity
[ Paper] 

Author Information

Linus Hamilton (MIT)
Frederic Koehler (MIT)
Ankur Moitra (MIT)

More from the Same Authors