A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf
.
Online Spectral Learning on a Graph with Bandit Feedback
2014
2014 IEEE International Conference on Data Mining
Online learning on a graph is appealing due to its efficiency. However, existing online learning algorithms on a graph are limited to binary classification. Moreover, they require accessing the full label information, where the label oracle needs to return the true class label after the learner makes classification of each node. In many application scenarios, we only have access to partial label information, where the label oracle will return a single bit indicating whether the prediction is
doi:10.1109/icdm.2014.72
dblp:conf/icdm/GuH14
fatcat:ni7mhvykozh2bn7rridxgmombm