Blar i NTNU Open på forfatter "He, Zhenli"
-
Energy-efficient computation offloading strategy with task priority in cloud assisted multi-access edge computing
He, Zhenli; Xu, Yanan; Liu, Di; Zhou, Wei; Li, Keqin (Peer reviewed; Journal article, 2023)Multi-access edge computing (MEC) provides cloud-like services at the edge of the radio access network close to mobile devices (MDs). This infrastructure can provide low-latency services to MDs and significantly reduce the ... -
Improving robustness of convolutional neural networks using element-wise activation scaling
Zhang, Zhi-Yuan; Ren, Hao; He, Zhenli; Zhou, Wei; Liu, Di (Peer reviewed; Journal article, 2023)Recent works reveal that re-calibrating intermediate activation of adversarial examples can improve the adversarial robustness of CNN models. The state of the arts exploit this feature at the channel level to help CNN ... -
OCAP: On-device Class-Aware Pruning for personalized edge DNN models
Ma, Ye-Da; Zhao, Zhi-Chao; Liu, Di; He, Zhenli; Zhou, Wei (Journal article; Peer reviewed, 2023)In this paper, we propose a new on-device class-aware pruning method for edge systems, namely OCAP. The motivation behind is that Deep Neural Network (DNN) models are usually trained with a large dataset so that they can ...