1 code implementation • 11 Jan 2024 • Wujie Sun, Defang Chen, Jiawei Chen, Yan Feng, Chun Chen, Can Wang
Deep learning has witnessed significant advancements in recent years at the cost of increasing training, inference, and model storage overhead.
2 code implementations • 30 Nov 2023 • Zhenyu Zhou, Defang Chen, Can Wang, Chun Chen
Sampling from diffusion models can be treated as solving the corresponding ordinary differential equations (ODEs), with the aim of obtaining an accurate solution with as few number of function evaluations (NFE) as possible.
1 code implementation • 10 Jul 2023 • Shiya Luo, Defang Chen, Can Wang
Existing works generally synthesize data from the pre-trained teacher model to replace the original training data for student learning.
1 code implementation • 11 Jun 2023 • Hailin Zhang, Defang Chen, Can Wang
Multi-Teacher knowledge distillation provides students with additional supervision from multiple pre-trained teachers with diverse information sources.
no code implementations • 31 May 2023 • Defang Chen, Zhenyu Zhou, Jian-Ping Mei, Chunhua Shen, Chun Chen, Can Wang
Recent years have witnessed significant progress in developing effective training and fast sampling techniques for diffusion models.
1 code implementation • 22 Nov 2022 • Wujie Sun, Defang Chen, Can Wang, Deshi Ye, Yan Feng, Chun Chen
Instead of aligning output images, we distill teacher's sharpened feature distribution into the student with a dataset-independent classifier, making the student focus on those important features to improve performance.
no code implementations • 25 Oct 2022 • Jiongyu Guo, Defang Chen, Can Wang
Alignahead++ transfers structure and feature information in a student layer to the previous layer of another simultaneously trained student model in an alternating training procedure.
no code implementations • 7 Aug 2022 • Junkun Yuan, Xu Ma, Defang Chen, Kun Kuang, Fei Wu, Lanfen Lin
To escape from the dilemma between domain generalization and annotation costs, in this paper, we introduce a novel task named label-efficient domain generalization (LEDG) to enable model generalization with label-limited source domains.
no code implementations • 7 Jun 2022 • Zhehui Zhou, Defang Chen, Can Wang, Yan Feng, Chun Chen
Iteratively incorporating and accumulating iteration-based semantic information enables the low-dimensional model to be more expressive for better link prediction in KGs.
1 code implementation • 5 May 2022 • Jiongyu Guo, Defang Chen, Can Wang
Existing knowledge distillation methods on graph neural networks (GNNs) are almost offline, where the student model extracts knowledge from a powerful teacher model to improve its performance.
1 code implementation • CVPR 2022 • Defang Chen, Jian-Ping Mei, Hailin Zhang, Can Wang, Yan Feng, Chun Chen
Knowledge distillation aims to compress a powerful yet cumbersome teacher model into a lightweight student model without much sacrifice of performance.
Ranked #3 on Knowledge Distillation on CIFAR-100
no code implementations • 16 Feb 2022 • Shiya Luo, Defang Chen, Can Wang
Knowledge distillation aims to enhance the performance of a lightweight student model by exploiting the knowledge from a pre-trained cumbersome teacher model.
1 code implementation • 30 Dec 2021 • Hailin Zhang, Defang Chen, Can Wang
Knowledge distillation is initially introduced to utilize additional supervision from a single teacher model for the student model training.
no code implementations • 28 Dec 2021 • Can Wang, Zhe Wang, Defang Chen, Sheng Zhou, Yan Feng, Chun Chen
However, its effect on graph neural networks is less than satisfactory since the graph topology and node attributes are likely to change in a dynamic way and in this case a static teacher model is insufficient in guiding student training.
1 code implementation • 13 Oct 2021 • Junkun Yuan, Xu Ma, Defang Chen, Fei Wu, Lanfen Lin, Kun Kuang
Domain generalization (DG) aims to learn from multiple known source domains a model that can generalize well to unknown target domains.
1 code implementation • 2 Oct 2021 • Junkun Yuan, Xu Ma, Defang Chen, Kun Kuang, Fei Wu, Lanfen Lin
In this paper, we investigate a Single Labeled Domain Generalization (SLDG) task with only one source domain being labeled, which is more practical and challenging than the CDG task.
no code implementations • 14 Sep 2021 • Defang Chen, Can Wang, Yan Feng, Chun Chen
Knowledge distillation is a generalized logits matching technique for model compression.
1 code implementation • ICCV 2021 • Sheng Zhou, Yucheng Wang, Defang Chen, Jiawei Chen, Xin Wang, Can Wang, Jiajun Bu
The holistic knowledge is represented as a unified graph-based embedding by aggregating individual knowledge from relational neighborhood samples with graph neural networks, the student network is learned by distilling the holistic knowledge in a contrastive manner.
2 code implementations • 6 Dec 2020 • Defang Chen, Jian-Ping Mei, Yuan Zhang, Can Wang, Yan Feng, Chun Chen
Knowledge distillation is a technique to enhance the generalization ability of a student model by exploiting outputs from a teacher model.
no code implementations • 2 Oct 2020 • Zheng Li, Ying Huang, Defang Chen, Tianren Luo, Ning Cai, Zhigeng Pan
Extensive experiments proved that our method significantly enhances the diversity among student models and brings better distillation performance.
2 code implementations • 1 Dec 2019 • Defang Chen, Jian-Ping Mei, Can Wang, Yan Feng, Chun Chen
The second-level distillation is performed to transfer the knowledge in the ensemble of auxiliary peers further to the group leader, i. e., the model used for inference.