Metraipolis › archive › dc_f133f9e4694e

Learning the Robustness Mechanism with Bilevel Optimization — arXiv:2609.35541

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.35541
License (as archived)
CC0-1.0
Archived
2026-10-07T22:39:13.418Z
SHA-256 of the archived text
2f0a93601254881721ac32b8c40aa2779807305241337bbbf2294302eef3d194
Stable URL of this page
https://metraipolis.com/documento/dc_f133f9e4694e

How to cite

Learning the Robustness Mechanism with Bilevel Optimization — arXiv:2609.35541 — arXiv (en), https://arxiv.org/abs/2609.35541 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f133f9e4694e (Metraipolis archive dc_f133f9e4694e, sha256 2f0a93601254881721ac32b8c40aa2779807305241337bbbf2294302eef3d194)
BibTeX
@misc{dcf133f9e4694e,
  title = {Learning the Robustness Mechanism with Bilevel Optimization — arXiv:2609.35541},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.35541},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f133f9e4694e; sha256 2f0a93601254881721ac32b8c40aa2779807305241337bbbf2294302eef3d194},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.35541. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We propose a distributionally robust learning framework where parameters defining the robustness mechanism are learned from held-out data instead of extensively tuned. Using bilevel optimization with both upper and lower level minimax problems, we create two instances of our framework to tackle setups with and without group labels in the training set. Theoretically, we provide sample complexity analysis for our robustness mechanism learning paradigm, showing that it achieves generalization guarantees comparable to exhaustive grid search while being more computationally efficient. Empirically, we evaluate our framework under a challenging setup when both intra-group and inter-group test distribution shifts occur at the same time, thereby demonstrating the efficacy and scalability of our method.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.35541 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T22:14:28.471Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).