TIDOT: A Teacher Imitation Learning Approach for Domain Adaptation with Optimal Transport

Nguyen, Tuan; Le, Trung; Dam, Nhan; Tran, Quan Hung; Nguyen, Truyen; Phung, Dinh

doi:10.24963/ijcai.2021/394

TIDOT: A Teacher Imitation Learning Approach for Domain Adaptation with Optimal Transport

Tuan Nguyen, Trung Le, Nhan Dam, Quan Hung Tran, Truyen Nguyen, Dinh Phung

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 2862-2868.

https://doi.org/10.24963/ijcai.2021/394

Keywords

Machine Learning: Transfer, Adaptation, Multi-task Learning

Abstract

Using the principle of imitation learning and the theory of optimal transport we propose in this paper a novel model for unsupervised domain adaptation named Teacher Imitation Domain Adaptation with Optimal Transport (TIDOT). Our model includes two cooperative agents: a teacher and a student. The former agent is trained to be an expert on labeled data in the source domain, whilst the latter one aims to work with unlabeled data in the target domain. More specifically, optimal transport is applied to quantify the total of the distance between embedded distributions of the source and target data in the joint space, and the distance between predictive distributions of both agents, thus by minimizing this quantity TIDOT could mitigate not only the data shift but also the label shift. Comprehensive empirical studies show that TIDOT outperforms existing state-of-the-art performance on benchmark datasets.

Tuan Nguyen, Trung Le, Nhan Dam, Quan Hung Tran, Truyen Nguyen, and Dinh Phung. 2021. TIDOT: A Teacher Imitation Learning Approach for Domain Adaptation with Optimal Transport. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 2862–2868. https://doi.org/10.24963/ijcai.2021/394

@inproceedings{ijcai2021p394,
  title     = {TIDOT: A Teacher Imitation Learning Approach for Domain Adaptation with Optimal Transport},
  author    = {Nguyen, Tuan and Le, Trung and Dam, Nhan and Tran, Quan Hung and Nguyen, Truyen and Phung, Dinh},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Prof. Zhi-Hua Zhou},
  pages     = {2862--2868},
  year      = {2021},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2021/394},
  url       = {https://doi.org/10.24963/ijcai.2021/394},
}

TY  - CPAPER
TI  - TIDOT: A Teacher Imitation Learning Approach for Domain Adaptation with Optimal Transport
AU  - Nguyen, Tuan
AU  - Le, Trung
AU  - Dam, Nhan
AU  - Tran, Quan Hung
AU  - Nguyen, Truyen
AU  - Phung, Dinh
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 2862
EP  - 2868
DO  - 10.24963/ijcai.2021/394
UR  - https://doi.org/10.24963/ijcai.2021/394
ER  -