paper-with-me

홈 › Papers

On Certifying and Improving Generalization to Unseen Domains

2022-06-24 · Akshay Mehra, Bhavya Kailkhura, Pin-Yu Chen, Jihun Hamm

Domain Generalization (DG) aims to learn models whose performance remains high on unseen domains encountered at test-time by using data from multiple related source domains. Many existing DG algorithms reduce the divergence between source distributions in a representation space to potentially align the unseen domain close to the sources. This is motivated by the analysis that explains generalization to unseen domains using distributional distance (such as the Wasserstein distance) to the sources. However, due to the openness of the DG objective, it is challenging to evaluate DG algorithms comprehensively using a few benchmark datasets. In particular, we demonstrate that the accuracy of the models trained with DG methods varies significantly across unseen domains, generated from popular benchmark datasets. This highlights that the performance of DG methods on a few benchmark datasets may not be representative of their performance on unseen domains in the wild. To overcome this roadblock, we propose a universal certification framework based on distributionally robust optimization (DRO) that can efficiently certify the worst-case performance of any DG method. This enables a data-independent evaluation of a DG method complementary to the empirical evaluations on benchmark datasets. Furthermore, we propose a training algorithm that can be used with any DG method to provably improve their certified performance. Our empirical evaluation demonstrates the effectiveness of our method at significantly improving the worst-case loss (i.e., reducing the risk of failure of these models in the wild) without incurring a significant performance drop on benchmark datasets.

📄 PDF Abstract BibTeX arXiv:2206.12364

Code (1)

akshaymehra24/CertifiableDG 공식 구현 tf

Tasks

Domain Generalization

Methods 이 논문이 사용한 방법론

ALIGN In the ALIGN method, visual and language representations are jointly trained from noisy image alt-text data. The image and text encoders are learned via contrastive loss…

Similar Papers 제목 키워드 기반

Certificate-Guided Evaluation of Reinforcement Learning Generalization

2026-05-30 · Vignesh Subramanian, Đorđe Žikelić, Suguman Bansal arxiv

This work presents a logic-driven framework to evaluate the performance of reinforcement learning (RL) algorithms in their ability to generalize to unseen tasks. Our framework defines a family of inductive reach-avoid ta…

Reinforcement Learning

Domain Generalization in Biosignal Classification

2020-11-12 · Theekshana Dissanayake, Tharindu Fernando, Simon Denman, Houman Ghaemmaghami 외

Objective: When training machine learning models, we often assume that the training data and evaluation data are sampled from the same distribution. However, this assumption is violated when the model is evaluated on ano…

ClassificationDomain AdaptationDomain GeneralizationGeneral Classification

INDIGO: Intrinsic Multimodality for Domain Generalization

2022-06-13 · Puneet Mangla, Shivam Chandhok, Milan Aggarwal, Vineeth N Balasubramanian 외

For models to generalize under unseen domains (a.k.a domain generalization), it is crucial to learn feature representations that are domain-agnostic and capture the underlying semantics that makes up an object category. …

Domain Generalization

Certifying Out-of-Domain Generalization for Blackbox Functions

2022-02-03 · Maurice Weber, Linyi Li, Boxin Wang, Zhikuan Zhao 외

Certifying the robustness of model performance under bounded data distribution drifts has recently attracted intensive interest under the umbrella of distributional robustness. However, existing techniques either make st…

Domain Generalization

Exploring Dualistic Meta-Learning to Enhance Domain Generalization in Open Set Scenarios

2026-06-22 · Xiran Wang, Jian Zhang, Lei Qi, Yang Gao 외 arxiv

Domain generalization learns from multiple source domains to generalize to unseen target domains. However, it often neglects the realistic case of label mismatch between source and target. Open set domain generalization …

Domain Generalization