Loughborough University

File(s) under embargo

Reason: This conference paper will be made available when published.

Single domain generalization via unsupervised diversity probe

conference contribution
posted on 2023-08-21, 15:54 authored by Kehua Guo, Rui Ding, Tian Qiu, Xiangyuan Zhu, Zheng Wu, Liwei Wang, Hui FangHui Fang

Single domain generalization (SDG) is a realistic yet challenging domain generalization scenario that aims to generalize a model trained on a single domain to multiple unseen domains. Typical SDG methods are essentially supervised data augmentation strategies, which tend to enhance the novelty rather than the diversity of augmented samples. Insufficient diversity may jeopardize the model generalization ability. In this paper, we propose a novel adversarial method, termed Unsupervised Diversity Probe (UDP), to synthesize novel and diverse samples in fully unsupervised settings. More specifically, to ensure that samples are novel, we study SDG from an information-theoretic perspective that minimizes the uncertainty coefficients between synthesized and source samples. Considering that the variation in a single source domain is limited, we introduce a regularization imposed on the auxiliary module that synthesizes variable samples, incorporated with uncertainty coefficients in an adversarial manner to complement the diversity. Subsequently, an available region is utilized to guarantee the samples’ safety. For the network architecture, we design a simple probe module that can synthesize samples in several different aspects. UDP is an unsupervised and easy-to-implement method that solves SDG using only synthetic (source) samples, thus reducing the dependence on task models. Extensive experiments on three benchmark datasets show that UDP achieves remarkable results and outperforms existing supervised and unsupervised methods by a large margin in single domain generalization. Source code is available: https://github.com/Ruiding1/Diversity_Probe.


111 Project (No. D23006)

Natural Science Foundation of China under Grant 62076255 and 62102458

Open Research Projects of Zhejiang Lab (No. 2022RC0AB07)

Hunan Provincial Science and Technology Plan Project 2020SK2059

Key projects of Hunan Education Department 20A88

National Science Foundation of Hunan Province 2021JJ30082 and 2022JJ40640



  • Science


  • Computer Science

Published in

Proceedings of the 31st ACM International Conference on Multimedia (MM ’23)


31st ACM International Conference on Multimedia (MM ’23)


Association for Computing Machinery (ACM)


  • AM (Accepted Manuscript)

Rights holder

© Owner/Author(s)

Publisher statement

© Owner/Author | 2023. This is the author's version of the work. It is posted here for your personal use. Not for redistribution. The definitive Version of Record was published in Proceedings of the 31st ACM International Conference on Multimedia (MM ’23), https://doi.org/10.1145/3581783.3612375.

Acceptance date


Copyright date



  • en


Ottawa, Canada

Event dates

29th October 2023 - 3rd November 2023


Dr Hui Fang. Deposit date: 3 August 2023

Usage metrics

    Loughborough Publications


    No categories selected