Download PDF

International Joint Conference on Artificial Intelligence, Date: 2016/07/09 - 2016/07/15, Location: New York, NY

Publication date: 2016-07-01
Volume: 2016-January Pages: 2082 - 2088
ISSN: 978-1-57735-770-4
Publisher: AAAI Press

Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence

Author:

Verachtert, Aäron
Blockeel, Hendrik ; Davis, Jesse ; Kambhampati, Subbarao

Keywords:

Energy efficient machine learning, naive Bayes

Abstract:

Energy efficiency is a concern for any software running on mobile devices. As such software employs machine-learned models to make predictions, this motivates research on efficiently executable models. In this paper, we propose a variant of the widely used Naive Bayes (NB) learner that yields a more efficient predictive model. In contrast to standard NB, where the learned model inspects all features to come to a decision, or NB with feature selection, where the model uses a fixed subset of the features, our model dynamically determines, on a case-by-case basis, when to stop inspecting features. We show that our approach is often much more efficient than the current state of the art, without loss of accuracy.