We make a small step towards extending Rate-Distortion Theory to channels subject to scaling and occlusion phenomena, ubiquitous in remote sensing and imaging data. We focus on the simplest possible decision task (binary classification), and on the simplest possible data formation model (“cartoon flatland”). We show that, in order to trade off the error rate, it is necessary to exercise control on the sensing process, and we characterize such a tradeoff.
Controlled Recognition Bounds for Scaling and Occlusion Channels
CHIUSO, ALESSANDRO
2011
Abstract
We make a small step towards extending Rate-Distortion Theory to channels subject to scaling and occlusion phenomena, ubiquitous in remote sensing and imaging data. We focus on the simplest possible decision task (binary classification), and on the simplest possible data formation model (“cartoon flatland”). We show that, in order to trade off the error rate, it is necessary to exercise control on the sensing process, and we characterize such a tradeoff.File in questo prodotto:
Non ci sono file associati a questo prodotto.
Pubblicazioni consigliate
Caricamento pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
