Cascaded Classifier for Pareto-Optimal Accuracy-Cost Trade-Off Using off-the-Shelf ANNs
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.
Keywords
Cite
@article{arxiv.2110.14256,
title = {Cascaded Classifier for Pareto-Optimal Accuracy-Cost Trade-Off Using off-the-Shelf ANNs},
author = {Cecilia Latotzke and Johnson Loh and Tobias Gemmeke},
journal= {arXiv preprint arXiv:2110.14256},
year = {2022}
}
Comments
The final accepted publication was presented on the 7th International Conference on Machine Learning, Optimization, Data Science (LOD), October 4 - 8, 2021 in Grasmere, Lake District, England