Multiple-Instance Learning of Real-Valued Data

Abstract

The multiple-instance learning model has received much attention recently with a primary application area being that of drug activity prediction. Most prior work on multiple-instance learning has been for concept learning, yet for drug activity prediction, the label is a real-valued affinity measurement giving the binding strength. We present extensions of k-nearest neighbors (k-NN), Citation-kNN, and the diverse density algorithm for the real-valued setting and study their performance on Boolean and real-valued data. We also provide a method for generating chemically realistic artificial data. 1.

Cite

Text

Amar et al. "Multiple-Instance Learning of Real-Valued Data." International Conference on Machine Learning, 2001. doi:10.5555/944919.944949

Markdown

[Amar et al. "Multiple-Instance Learning of Real-Valued Data." International Conference on Machine Learning, 2001.](https://mlanthology.org/icml/2001/amar2001icml-multiple/) doi:10.5555/944919.944949

BibTeX

@inproceedings{amar2001icml-multiple,
  title     = {{Multiple-Instance Learning of Real-Valued Data}},
  author    = {Amar, Robert A. and Dooly, Daniel R. and Goldman, Sally A. and Zhang, Qi},
  booktitle = {International Conference on Machine Learning},
  year      = {2001},
  pages     = {3-10},
  doi       = {10.5555/944919.944949},
  url       = {https://mlanthology.org/icml/2001/amar2001icml-multiple/}
}