Delta divergence: A novel decision cognizant measure of classifier incongruence
This work addresses a specific issue in pattern recognition for detecting classifier incongruence, but it appears incremental as it builds on existing divergence methods with a novel focus.
The paper tackles the problem of measuring disagreement between classifiers, which can indicate anomalies, by proposing a new divergence measure called Delta divergence that is decision cognizant and focuses on dominant hypotheses to reduce clutter, especially with many classes, and shows experimental superiority over baseline measures.
Disagreement between two classifiers regarding the class membership of an observation in pattern recognition can be indicative of an anomaly and its nuance. As in general classifiers base their decision on class aposteriori probabilities, the most natural approach to detecting classifier incongruence is to use divergence. However, existing divergences are not particularly suitable to gauge classifier incongruence. In this paper, we postulate the properties that a divergence measure should satisfy and propose a novel divergence measure, referred to as Delta divergence. In contrast to existing measures, it is decision cognizant. The focus in Delta divergence on the dominant hypotheses has a clutter reducing property, the significance of which grows with increasing number of classes. The proposed measure satisfies other important properties such as symmetry, and independence of classifier confidence. The relationship of the proposed divergence to some baseline measures is demonstrated experimentally, showing its superiority.