Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining 2014
DOI: 10.1145/2623330.2623760
|View full text |Cite
|
Sign up to set email alerts
|

Active semi-supervised learning using sampling theory for graph signals

Abstract: We consider the problem of offline, pool-based active semisupervised learning on graphs. This problem is important when the labeled data is scarce and expensive whereas unlabeled data is easily available. The data points are represented by the vertices of an undirected graph with the similarity between them captured by the edge weights. Given a target number of nodes to label, the goal is to choose those nodes that are most informative and then predict the unknown labels. We propose a novel framework for this … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
123
0

Year Published

2016
2016
2022
2022

Publication Types

Select...
5
3
2

Relationship

1
9

Authors

Journals

citations
Cited by 127 publications
(123 citation statements)
references
References 23 publications
0
123
0
Order By: Relevance
“…where f c denotes the classical continuous signal, P c T denotes the time domain downsampling operator, I denotes the identity operator, P c w denotes the frequency domain cut-off operator, and According to the principle of Papoulis-Gerchberg Algorithm, an iterative least square reconstruction (ILSR) algorithm is proposed in [9,17] for the signal processing on graphs. At each iteration, ILSR resets the signal samples on the downsampling set S to the actual given samples and then projects the graph signal onto the lowpass filtering subspace.…”
Section: Reconstruction Of a Bandlimited Graph Signalmentioning
confidence: 99%
“…where f c denotes the classical continuous signal, P c T denotes the time domain downsampling operator, I denotes the identity operator, P c w denotes the frequency domain cut-off operator, and According to the principle of Papoulis-Gerchberg Algorithm, an iterative least square reconstruction (ILSR) algorithm is proposed in [9,17] for the signal processing on graphs. At each iteration, ILSR resets the signal samples on the downsampling set S to the actual given samples and then projects the graph signal onto the lowpass filtering subspace.…”
Section: Reconstruction Of a Bandlimited Graph Signalmentioning
confidence: 99%
“…After the nodes selected by each method have been sampled, we reconstruct the unknown label signal using the approximate POCS based bandlimited reconstruction scheme [8] to get the soft labels. We threshold these soft labels to get the final label predictions.…”
Section: Methodsmentioning
confidence: 99%
“…SOpt selects queries by which nodes have been labeled, ignoring what labels they have. In fact, this is the common characteristic of many graph-based active learning algorithms [6,7,8]. This is why SOpt is not able to optimize exploitation queries, which results in higher error than other methods as we show in toy experiments in Section 4.…”
Section: Introductionmentioning
confidence: 97%