Can I read RC2: Resource-efficient Collaborative Cognition Method for Internet-of-Things Devices on EtoBox?

RC2: Resource-efficient Collaborative Cognition Method for Internet-of-Things Devices by Yuanyuan Bao and Wai Chen is a scholarly article available to read on EtoBox.

What is RC2: Resource-efficient Collaborative Cognition Method for Internet-of-Things Devices about?

Given that cognitive capabilities of Internet-of-Things (IoT) devices often have huge differences, especially due to their deployments in highly personal environments, it is plausible that collaborating IoT devices for cognition can lead to higher accuracy. However, designing a collaborative cognition mechanism for resource-constrained IoT devices is challenging because the collaboration needs to involve a subset of IoT devices, yet still be accurate. Although existing methods (e.g., ensemble pruning) can help the models to ensemble without sacrificing accuracy, these approaches only consider accuracy without computational cost, which is especially important for resource-scarce IoT devices. In this paper, we propose RC 2 , a resource-efficient collaborative cognition approach for IoT devices collaboration. We propose to explicitly incorporate computational cost information into the main objective so that the approach can obtain a collaboration that achieves a good trade-off between accuracy and computational cost. We conduct empirical evaluation of our method compared with two state-of-the-art methods; and the evaluation shows the effectiveness and efficiency of our method. ## CCS

Author

Yuanyuan Bao and Wai Chen

Publisher

ACM

Published

2019

Language

EN