Allerton 2015 Paper Abstract


Paper ThD3.4

Karzand, Mina (MIT), Bresler, Guy (MIT)

Inferning Trees

Scheduled for presentation during the Regular Session "Machine Learning II" (ThD3), Thursday, October 1, 2015, 16:30−16:50, Butternut

53rd Annual Allerton Conference on Communication, Control, and Computing, Sept 29-Oct 2, 2015, Allerton Park and Retreat Center, Monticello, IL, USA

This information is tentative and subject to change. Compiled on December 5, 2021

Keywords Sparse Data Analysis, Universal Algorithms and Machine Learning, Detection and Estimation


We consider the problem of learning an Ising model with the goal of subsequently performing inference from partial observations. This is in contrast to most other work on graphical model learning, which tries to learn either the true underlying graph or a model that is close in KL-divergence. These objectives require a lower bound on the strength of edges for identifiability of the model. We show that in the relatively simple case of tree models, the Chow-Liu algorithm learns a distribution with accurate low-order marginals despite the model being non-identifiable. In other words, a model that appears rather different from the truth nevertheless allows to carry out inference accurately.



All Content © PaperCept, Inc..

This site is protected by copyright and trademark laws under US and International law.
All rights reserved. © 2002-2021 PaperCept, Inc.
Page generated 2021-12-05  10:04:01 PST  Terms of use