Bayesian Optimal Active Search on Graphs

  title={Bayesian Optimal Active Search on Graphs},
  author={Roman Garnett and Yamuna Krishnamurthy and Donghan Wang and Jeff Schneider},
In many classification problems, including numerous examples on modern large-scale graph datasets, a large quantity of unlabeled data are available. The cost of obtaining a label for such data can be very expensive, for example when human intervention is required. Both the semi-supervised and active learning communities have approached such problems. The former focuses on how to aid the classification task by exploiting the distribution of unlabeled data, and the latter addresses the problem of… CONTINUE READING

Similar Papers

Loading similar papers…