Sifan Long
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Sifan Long has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Shengsheng Wang
College of Computer Science and Technology, Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education, Jilin University, Changchun, China
5 shared publications - Bilin Wang
College of Computer Science and Technology, Jilin University, Changchun, China
3 shared publications - Fu Zihao
College of Computer Science and Technology, Jilin University, Changchun, China
3 shared publications - Xin Zhao
State Key Laboratory of Deep Earth Exploration and Imaging and the College of Instrumentation Science and Electrical Engineering, Jilin University, Changchun, China
3 shared publications - Hao Chai
Hunan Engineering Research Center of Advanced Embedded Computing and Intelligent Medical Systems, Xiangnan University, Chenzhou, China
2 shared publications - Yiyang Li
College of Computer Science and Technology, Jilin University, Changchun, China
2 shared publications
Latest Publications
- 27 November 2023
- Published by Springer Nature in The Visual Computer
- Vol. 40 (9) , 6261-6277
Bidirectional feature enhancement transformer for unsupervised domain adaptation
- 27 November 2023
- Published by Springer Nature in The Visual Computer
- Vol. 40 (9) , 6261-6277
Bidirectional feature enhancement transformer for unsupervised domain adaptation
- 9 May 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (30) , 47347-47366
Causal view mechanism for adversarial domain adaptation
- 9 May 2023
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 82 (30) , 47347-47366
Causal view mechanism for adversarial domain adaptation
- Open Access
- 8 May 2023
- Published by Springer Nature in Research Square
Open set domain adaptive method based on energy separation strategy
Open Set Domain Adaptation (OSDA) aims to reduce the variation across domains while distinguishing between known samples and unknown samples. However, existing OSDA methods are based on deep neural network classifiers to separate unknown samples, which leads the network to produce overconfident predictions and fails to establish the boundary between the known and the unknown. We propose an Energy-based Open Set Domain Adaptation method (EOS). Specifically, EOS is a new two-stage approach of separation followed by alignment. We use an energy-based anomaly detection strategy as a separation method for unknown samples, transforming the traditional K-way classification task into a K+1-dimensional classifier that uses an additional dimension to model the uncertainty of Out-of-distribution (OOD) samples. Then, we use a coarse-to-fine separation method to continuously adjust the separation results and add the exact separation results as weights in the alignment process, weighing their importance to the feature distribution alignment. In the alignment phase we also optimize our separation network module at the same time, so that the module can be better adapted with invariant features. We have done experiments on the standard ground Office-31, Office-Home, and VisDA-2017 benchmarks, and the results show that our approach outperforms our competitors in most cases.
- 8 May 2023
- Published by Springer Nature in Research Square
Open AccessOpen set domain adaptive method based on energy separation strategy
Open Set Domain Adaptation (OSDA) aims to reduce the variation across domains while distinguishing between known samples and unknown samples. However, existing OSDA methods are based on deep neural network classifiers to separate unknown samples, which leads the network to produce overconfident predictions and fails to establish the boundary between the known and the unknown. We propose an Energy-based Open Set Domain Adaptation method (EOS). Specifically, EOS is a new two-stage approach of separation followed by alignment. We use an energy-based anomaly detection strategy as a separation method for unknown samples, transforming the traditional K-way classification task into a K+1-dimensional classifier that uses an additional dimension to model the uncertainty of Out-of-distribution (OOD) samples. Then, we use a coarse-to-fine separation method to continuously adjust the separation results and add the exact separation results as weights in the alignment process, weighing their importance to the feature distribution alignment. In the alignment phase we also optimize our separation network module at the same time, so that the module can be better adapted with invariant features. We have done experiments on the standard ground Office-31, Office-Home, and VisDA-2017 benchmarks, and the results show that our approach outperforms our competitors in most cases.
- 9 February 2023
- Published by Springer Nature in Applied Intelligence
- Vol. 53 (15) , 18790-18805
Sample separation and domain alignment complementary learning mechanism for open set domain adaptation
- 9 February 2023
- Published by Springer Nature in Applied Intelligence
- Vol. 53 (15) , 18790-18805
Sample separation and domain alignment complementary learning mechanism for open set domain adaptation
- 1 April 2022
- Published by Springer Nature in Applied Intelligence
- Vol. 52 (15) , 17326-17340
Cross-domain feature enhancement for unsupervised domain adaptation
- 1 April 2022
- Published by Springer Nature in Applied Intelligence
- Vol. 52 (15) , 17326-17340
Cross-domain feature enhancement for unsupervised domain adaptation