TY - CONF
T1 - Semi-supervised learning with competitive infection models
AU - Rosenfeld, Nir
AU - Globerson, Amir
N1 - Publisher Copyright:
Copyright 2018 by the author(s).
PY - 2018
Y1 - 2018
N2 - The goal in semi-supervised learning is to effectively combine labeled and unlabeled data. One way to do this is by encouraging smoothness across edges in a graph whose nodes correspond to input examples. In many graph-based methods, labels can be thought of as propagating over the graph, where the underlying propagation mechanism is based on random walks or on averaging dynamics. While theoretically elegant, these dynamics suffer from several drawbacks which can hurt predictive performance. Our goal in this work is to explore alternative mechanisms for propagating labels. In particular, we propose a method based on dynamic infection processes, where unlabeled nodes can be “infected” with the label of their already infected neighbors. Our algorithm is efficient and scalable, and an analysis of the underlying optimization objective reveals a surprising relation to other Laplacian approaches. We conclude with a thorough set of experiments across multiple benchmarks and various learning settings.
AB - The goal in semi-supervised learning is to effectively combine labeled and unlabeled data. One way to do this is by encouraging smoothness across edges in a graph whose nodes correspond to input examples. In many graph-based methods, labels can be thought of as propagating over the graph, where the underlying propagation mechanism is based on random walks or on averaging dynamics. While theoretically elegant, these dynamics suffer from several drawbacks which can hurt predictive performance. Our goal in this work is to explore alternative mechanisms for propagating labels. In particular, we propose a method based on dynamic infection processes, where unlabeled nodes can be “infected” with the label of their already infected neighbors. Our algorithm is efficient and scalable, and an analysis of the underlying optimization objective reveals a surprising relation to other Laplacian approaches. We conclude with a thorough set of experiments across multiple benchmarks and various learning settings.
UR - http://www.scopus.com/inward/record.url?scp=85060561111&partnerID=8YFLogxK
M3 - ???researchoutput.researchoutputtypes.contributiontoconference.paper???
AN - SCOPUS:85060561111
SP - 336
EP - 346
T2 - 21st International Conference on Artificial Intelligence and Statistics, AISTATS 2018
Y2 - 9 April 2018 through 11 April 2018
ER -