Search

Your search keyword '"Tang, Jialiang"' showing total 7 results
Start Over Author "Tang, Jialiang" Topic convolutional neural networks
7 results on '"Tang, Jialiang"'

2. Consistent Knowledge Distillation Based on Siamese Networks

Author

Tang, Jialiang, Yang, Xiaoyan, Cheng, Xin, Jiang, Ning, Yu, Wenxin, Zhang, Peng, Filipe, Joaquim, Editorial Board Member, Ghosh, Ashish, Editorial Board Member, Prates, Raquel Oliveira, Editorial Board Member, Zhou, Lizhu, Editorial Board Member, Mantoro, Teddy, editor, Lee, Minho, editor, Ayu, Media Anugerah, editor, Wong, Kok Wai, editor, and Hidayanto, Achmad Nizar, editor

Published
2021
Full Text
View/download PDF

3. Triplet Mapping for Continuously Knowledge Distillation

Author

Yang, Xiaoyan, Tang, Jialiang, Jiang, Ning, Yu, Wenxin, Zhang, Peng, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Mantoro, Teddy, editor, Lee, Minho, editor, Ayu, Media Anugerah, editor, Wong, Kok Wai, editor, and Hidayanto, Achmad Nizar, editor

Published
2021
Full Text
View/download PDF

4. Improving Shallow Neural Networks via Local and Global Normalization

Author

Jiang, Ning, Tang, Jialiang, Yang, Xiaoyan, Yu, Wenxin, Zhang, Peng, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Mantoro, Teddy, editor, Lee, Minho, editor, Ayu, Media Anugerah, editor, Wong, Kok Wai, editor, and Hidayanto, Achmad Nizar, editor

Published
2021
Full Text
View/download PDF

5. Local Feature Normalization

Author

Jiang, Ning, Tang, Jialiang, Yu, Wenxin, Zhou, Jinjia, Goos, Gerhard, Founding Editor, Hartmanis, Juris, Founding Editor, Bertino, Elisa, Editorial Board Member, Gao, Wen, Editorial Board Member, Steffen, Bernhard, Editorial Board Member, Woeginger, Gerhard, Editorial Board Member, Yung, Moti, Editorial Board Member, Qiu, Han, editor, Zhang, Cheng, editor, Fei, Zongming, editor, Qiu, Meikang, editor, and Kung, Sun-Yuan, editor

Published
2021
Full Text
View/download PDF

6. Search-and-Train: Two-Stage Model Compression and Acceleration

Author

Jiang, Ning, Tang, Jialiang, Zhang, Zhiqiang, Yu, Wenxin, Deng, Xin, Zhou, Jinjia, Filipe, Joaquim, Editorial Board Member, Ghosh, Ashish, Editorial Board Member, Prates, Raquel Oliveira, Editorial Board Member, Zhou, Lizhu, Editorial Board Member, Yang, Haiqin, editor, Pasupa, Kitsuchart, editor, Leung, Andrew Chi-Sing, editor, Kwok, James T., editor, Chan, Jonathan H., editor, and King, Irwin, editor

Published
2020
Full Text
View/download PDF

7. Positive-Unlabeled Learning for Knowledge Distillation.

Author

Jiang, Ning, Tang, Jialiang, and Yu, Wenxin

Subjects
IMAGE recognition (Computer vision), CONVOLUTIONAL neural networks, ARTIFICIAL intelligence, SUPERVISED learning, SECURE Sockets Layer (Computer network protocol), LEARNING, TEACHER training
Abstract

Convolutional neural networks (CNNs) have greatly promoted the development of artificial intelligence. In general, CNNs with high performance are over-parameterized, requiring massive calculations to process and predict the data. It leads CNNs unable to apply to exiting resource-limited intelligence devices. In this paper, we propose an efficient model compression framework based on knowledge distillation to train a compact student network by a large teacher network. Our key point is to introduce a positive-unlabeled (PU) classifier to promote the compressed student network to learn the features of the prominent teacher network as much as possible. During the training, the PU classifier is to discriminate the features of the teacher network as high-quality and discriminate the features of the student network as low-quality. Simultaneously, the student network learns knowledge from the teacher network through the soft-targets and attention features. Extensive experimental evaluations on four benchmark image classification datasets show that our method outperforms the prior works with a large margin at the same parameters and calculations cost. When selecting the VGGNet19 as the teacher network to train on the CIFAR dataset, the student network VGGNet13 achieves 94.47% and 75.73% accuracy on the CIFAR-10 and CIFAR-100 datasets, which improved 1.02% and 2.44%, respectively. [ABSTRACT FROM AUTHOR]

Published
2023
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results