CDF-Intervals: A Reliable Framework to Reason about Data with Uncertainty
Abstract
This research introduces a new constraint domain for reasoning about data with uncertainty. It extends convex modeling with the notion of p-box to gain additional quantifiable information on the data whereabouts. Unlike existing approaches, the p-box envelops an unknown probability instead of approximating its representation. The p-box bounds are uniform cumulative distribution functions (cdf) in order to employ linear computations in the probabilistic domain. The reasoning by means of p-box cdf-intervals is an interval computation which is exerted on the real domain then it is projected onto the cdf domain. This operation conveys additional knowledge represented by the obtained probabilistic bounds. Empirical evaluation shows that, with minimal overhead, the output solution set realizes a full enclosure of the data along with tighter bounds on its probabilistic distributions.
Cite
@article{arxiv.1405.4008,
title = {CDF-Intervals: A Reliable Framework to Reason about Data with Uncertainty},
author = {Aya Saad},
journal= {arXiv preprint arXiv:1405.4008},
year = {2014}
}
Comments
10 pages, 15 Postscript figures, uses new_tlp.cls and acmtrans.bst, full version of a paper accepted to be presented at the Doctoral Consortium of the 30th International Conference on Logic Programming (ICLP 2014), July 19-22, Vienna, Austria. arXiv admin note: substantial text overlap with arXiv:1405.2801