To fuse or not to fuse: Fuser versus best classifier
Conference
·
OSTI ID:658387
A sample from a class defined on a finite-dimensional Euclidean space and distributed according to an unknown distribution is given. The authors are given a set of classifiers each of which chooses a hypothesis with least misclassification error from a family of hypotheses. They address the question of choosing the classifier with the best performance guarantee versus combining the classifiers using a fuser. They first describe a fusion method based on isolation property such that the performance guarantee of the fused system is at least as good as the best of the classifiers. For a more restricted case of deterministic classes, they present a method based on error set estimation such that the performance guarantee of fusing all classifiers is at least as good as that of fusing any subset of classifiers.
- Research Organization:
- Oak Ridge National Lab., TN (United States)
- Sponsoring Organization:
- USDOE Office of Energy Research, Washington, DC (United States); Oak Ridge National Lab., TN (United States); Department of Defense, Washington, DC (United States)
- DOE Contract Number:
- AC05-96OR22464
- OSTI ID:
- 658387
- Report Number(s):
- ORNL/CP--97084; CONF-980412--; ON: DE98005063; BR: 213001000; KC0401030; 43WJ07101
- Country of Publication:
- United States
- Language:
- English
Similar Records
On Optimal Projective Fusers for Function Estimators
A fusion method that performs better than best sensor
Projective Method for Generic Sensor Fusion Problem
Conference
·
Tue Jun 22 00:00:00 EDT 1999
·
OSTI ID:7929
A fusion method that performs better than best sensor
Conference
·
Sat Feb 28 23:00:00 EST 1998
·
OSTI ID:654156
Projective Method for Generic Sensor Fusion Problem
Conference
·
Sun Aug 15 00:00:00 EDT 1999
·
OSTI ID:9148