Domain Generalisation with Domain Augmented Supervised Contrastive Learning (Student Abstract)

Hoang-Son Le, Rini Akmeliawati, Gustavo Carneiro

AAAI 2021 pp. 15821-15822

doi:10.1609/AAAI.V35I18.17907 /aaai/2021/le2021aaai-domain/

Abstract

Domain generalisation (DG) methods address the problem of domain shift, when there is a mismatch between the distributions of training and target domains. Data augmentation approaches have emerged as a promising alternative for DG. However, data augmentation alone is not sufficient to achieve lower generalisation errors. This project proposes a new method that combines data augmentation and domain distance minimisation to address the problems associated with data augmentation and provide a guarantee on the learning performance, under an existing framework. Empirically, our method outperforms baseline results on DG benchmarks.

PDF AAAI Semantic Scholar

Cite

Text

Le et al. "Domain Generalisation with Domain Augmented Supervised Contrastive Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2021. doi:10.1609/AAAI.V35I18.17907

Markdown

[Le et al. "Domain Generalisation with Domain Augmented Supervised Contrastive Learning (Student Abstract)." AAAI Conference on Artificial Intelligence, 2021.](https://mlanthology.org/aaai/2021/le2021aaai-domain/) doi:10.1609/AAAI.V35I18.17907

BibTeX

@inproceedings{le2021aaai-domain,
  title     = {{Domain Generalisation with Domain Augmented Supervised Contrastive Learning (Student Abstract)}},
  author    = {Le, Hoang-Son and Akmeliawati, Rini and Carneiro, Gustavo},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2021},
  pages     = {15821-15822},
  doi       = {10.1609/AAAI.V35I18.17907},
  url       = {https://mlanthology.org/aaai/2021/le2021aaai-domain/}
}