Occam's hammer: a link between randomized learning and multiple testing FDR control

Details Occam's hammer: a link between randomized learning and multiple testing FDR control Occam's hammer: a link between randomized learning and multiple testing FDR control

2006-08-29

arxiv.org/abs/math/0608713v1

Mathematics

Statistics Theory

13 pages -- conference communication type format

Scientific paper

We establish a generic theoretical tool to construct probabilistic bounds for algorithms where the output is a subset of objects from an initial pool of candidates (or more generally, a probability distribution on said pool). This general device, dubbed "Occam's hammer'', acts as a meta layer when a probabilistic bound is already known on the objects of the pool taken individually, and aims at controlling the proportion of the objects in the set output not satisfying their individual bound. In this regard, it can be seen as a non-trivial generalization of the "union bound with a prior'' ("Occam's razor''), a familiar tool in learning theory. We give applications of this principle to randomized classifiers (providing an interesting alternative approach to PAC-Bayes bounds) and multiple testing (where it allows to retrieve exactly and extend the so-called Benjamini-Yekutieli testing procedure).

Affiliated with

Also associated with

No associations

Rating

Occam's hammer: a link between randomized learning and multiple testing FDR control does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Occam's hammer: a link between randomized learning and multiple testing FDR control, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Occam's hammer: a link between randomized learning and multiple testing FDR control will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-334025