Optimal Online Prediction in Adversarial Environments
Abstract
In many prediction problems, including those that arise in computer security and computational finance, the process generating the data is best modelled as an adversary with whom the predictor competes. Even decision problems that are not inherently adversarial can be usefully modeled in this way, since the assumptions are sufficiently weak that effective prediction strategies for adversarial settings are very widely applicable.
Cite
Text
Bartlett. "Optimal Online Prediction in Adversarial Environments." International Conference on Algorithmic Learning Theory, 2010. doi:10.1007/978-3-642-16108-7_6Markdown
[Bartlett. "Optimal Online Prediction in Adversarial Environments." International Conference on Algorithmic Learning Theory, 2010.](https://mlanthology.org/alt/2010/bartlett2010alt-optimal/) doi:10.1007/978-3-642-16108-7_6BibTeX
@inproceedings{bartlett2010alt-optimal,
title = {{Optimal Online Prediction in Adversarial Environments}},
author = {Bartlett, Peter L.},
booktitle = {International Conference on Algorithmic Learning Theory},
year = {2010},
pages = {34},
doi = {10.1007/978-3-642-16108-7_6},
url = {https://mlanthology.org/alt/2010/bartlett2010alt-optimal/}
}