Cascaded Classifier for Pareto-Optimal Accuracy-Cost Trade-Off Using off-the-Shelf ANNs

TL;DR

This paper introduces a novel analysis and methodology for cascaded classifiers that optimizes the accuracy-cost trade-off, enabling significant cost reductions while maintaining or scaling accuracy levels.

Contribution

It provides the first analysis of pass-on criteria in cascaded classifiers and a methodology to maximize accuracy and efficiency, adaptable to any classifier.

Findings

Cost reduced by 1.32x with preserved accuracy

Cost scalable over two orders with graceful accuracy degradation

Applicable to any state-of-the-art classifier

Abstract

Machine-learning classifiers provide high quality of service in classification tasks. Research now targets cost reduction measured in terms of average processing time or energy per solution. Revisiting the concept of cascaded classifiers, we present a first of its kind analysis of optimal pass-on criteria between the classifier stages. Based on this analysis, we derive a methodology to maximize accuracy and efficiency of cascaded classifiers. On the one hand, our methodology allows cost reduction of 1.32x while preserving reference classifier's accuracy. On the other hand, it allows to scale cost over two orders while gracefully degrading accuracy. Thereby, the final classifier stage sets the top accuracy. Hence, the multi-stage realization can be employed to optimize any state-of-the-art classifier.