Translator Disclaimer
June 2005 Hierarchical testing designs for pattern recognition
Gilles Blanchard, Donald Geman
Ann. Statist. 33(3): 1155-1202 (June 2005). DOI: 10.1214/009053605000000174

Abstract

We explore the theoretical foundations of a “twenty questions” approach to pattern recognition. The object of the analysis is the computational process itself rather than probability distributions (Bayesian inference) or decision boundaries (statistical learning). Our formulation is motivated by applications to scene interpretation in which there are a great many possible explanations for the data, one (“background”) is statistically dominant, and it is imperative to restrict intensive computation to genuinely ambiguous regions.

The focus here is then on pattern filtering: Given a large set $\mathcal {Y}$ of possible patterns or explanations, narrow down the true one Y to a small (random) subset $\widehat{Y}\subset\mathcal{Y}$ of “detected” patterns to be subjected to further, more intense, processing. To this end, we consider a family of hypothesis tests for YA versus the nonspecific alternatives YAc. Each test has null type I error and the candidate sets $A\subset\mathcal{Y}$ are arranged in a hierarchy of nested partitions. These tests are then characterized by scope (|A|), power (or type II error) and algorithmic cost.

We consider sequential testing strategies in which decisions are made iteratively, based on past outcomes, about which test to perform next and when to stop testing. The set is then taken to be the set of patterns that have not been ruled out by the tests performed. The total cost of a strategy is the sum of the “testing cost” and the “postprocessing cost” (proportional to ||) and the corresponding optimization problem is analyzed. As might be expected, under mild assumptions good designs for sequential testing strategies exhibit a steady progression from broad scope coupled with low power to high power coupled with dedication to specific explanations. In the assumptions ensuring this property a key role is played by the ratio cost/power. These ideas are illustrated in the context of detecting rectangles amidst clutter.

Citation

Download Citation

Gilles Blanchard. Donald Geman. "Hierarchical testing designs for pattern recognition." Ann. Statist. 33 (3) 1155 - 1202, June 2005. https://doi.org/10.1214/009053605000000174

Information

Published: June 2005
First available in Project Euclid: 1 July 2005

zbMATH: 1072.62052
MathSciNet: MR2195632
Digital Object Identifier: 10.1214/009053605000000174

Subjects:
Primary: 62H30, 62L05, 68T10
Secondary: 62H15, 68T45, 90B40

Rights: Copyright © 2005 Institute of Mathematical Statistics

JOURNAL ARTICLE
48 PAGES


SHARE
Vol.33 • No. 3 • June 2005
Back to Top