Journal article · 2026 · Open access

Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization

Danni Peng, Sinno Jialin Pan

Artificial Intelligence, 356, 104544 · Published 23 April 2026 · DOI 10.1016/j.artint.2026.104544

Open full textCite this paper

In brief

“Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization” is a 2026 journal article by Danni Peng and Sinno Jialin Pan, published in Artificial Intelligence. OpenAlex counts 4 citations as of 17 November 2026. It is open access under a CC BY-NC licence: the full text is free to read at Artificial Intelligence.

Citations
4
Most cited in
2025 (3)
References
58
Authors
2

Abstract

To address distribution shifts between training and test data, domain generalization (DG) leverages multiple source domains to learn a model that generalizes well to unseen domains. However, existing DG methods often overfit to the source domains, partly due to the limited coverage of the expected region in feature space. Motivated by this, we propose performing mixup with data interpolation and extrapolation to cover potentially unseen regions. To prevent the detrimental effects of unconstrained extrapolation, we carefully design a policy to generate the instance weights, named F latness-aware G radient-based Mix up (FGMix). The policy relies on gradient-based compatibilities to assign greater weights to instances that carry more invariant information and learn the mixup policy towards flatter minima for better generalization. On the DomainBed benchmark, we validate the efficacy of various designs of FGMix and demonstrate its superiority over other DG algorithms.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Peng, D., & Pan, S. J. (2026). Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization. Artificial Intelligence, 356, 104544. https://doi.org/10.1016/j.artint.2026.104544

MLA 9

Peng, Danni, and Sinno Jialin Pan. “Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization.” Artificial Intelligence, vol. 356, 2026, p. 104544. https://doi.org/10.1016/j.artint.2026.104544.

Chicago

Peng, Danni and Sinno Jialin Pan. 2026. “Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization.” Artificial Intelligence 356: 104544. https://doi.org/10.1016/j.artint.2026.104544.

Harvard

Peng, D. and Pan, S. J. (2026) ‘Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization’, Artificial Intelligence, 356, p. 104544. doi:10.1016/j.artint.2026.104544.

IEEE

D. Peng and S. J. Pan, “Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization,” Artificial Intelligence, vol. 356, p. 104544, 2026, doi: 10.1016/j.artint.2026.104544.

BibTeX

@article{peng2026learning,
  title     = {{Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization}},
  author    = {Peng, Danni and Pan, Sinno Jialin},
  journal   = {Artificial Intelligence},
  year      = {2026},
  volume    = {356},
  pages     = {104544},
  doi       = {10.1016/j.artint.2026.104544},
  url       = {https://doi.org/10.1016/j.artint.2026.104544}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning

  • domain generalization
  • mixup
  • data extrapolation
  • data interpolation
  • instance weighting
  • distribution shift
  • invariant learning

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Peng, D., & Pan, S. J. (2026). Learning gradient-based mixup with extrapolation toward flatter minima for domain generalization. Artificial Intelligence, 356, 104544. https://doi.org/10.1016/j.artint.2026.104544

How many times has this paper been cited?

4 times, according to OpenAlex on 17 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Artificial Intelligence: sign in and open the full text from this page.

Who wrote this paper?

Danni Peng, Sinno Jialin Pan.

What is the DOI of this paper?

10.1016/j.artint.2026.104544, permanently at https://doi.org/10.1016/j.artint.2026.104544. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 17 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.