# Butterfly: One-step Approach towards Wildly Unsupervised Domain Adaptation

arXiv: https://arxiv.org/abs/1905.07720 · HTML: https://mlsanity.com/paper/1905.07720/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 May 2019.

## Built on by (most cited first, 8 of 8)

- [A Theory of Label Propagation for Subpopulation Shift](https://mlsanity.com/paper/2102.11203/) (2021)
- [Meta Discovery: Learning to Discover Novel Classes given Very Limited Data](https://mlsanity.com/paper/2102.04002/) (2021)
- [Learning Bounds for Open-Set Learning](https://mlsanity.com/paper/2106.15792/) (2021)
- [TOHAN: A One-step Approach towards Few-shot Hypothesis Adaptation](https://mlsanity.com/paper/2106.06326/) (2021)
- [GearNet: Stepwise Dual Learning for Weakly Supervised Domain Adaptation](https://mlsanity.com/paper/2201.06001/) (2022)
- [Towards Robust Cross-domain Image Understanding with Unsupervised Noise Removal](https://mlsanity.com/paper/2109.04284/) (2021)
- [VALERIAN: Invariant Feature Learning for IMU Sensor-based Human Activity Recognition in the Wild](https://mlsanity.com/paper/2303.06048/) (2023)
- [Learning to Discover Knowledge: A Weakly-Supervised Partial Domain Adaptation Approach](https://mlsanity.com/paper/2406.14274/) (2024)

## Abstract (arXiv)

In unsupervised domain adaptation (UDA), classifiers for the target domain (TD) are trained with clean labeled data from the source domain (SD) and unlabeled data from TD. However, in the wild, it is difficult to acquire a large amount of perfectly clean labeled data in SD given limited budget. Hence, we consider a new, more realistic and more challenging problem setting, where classifiers have to be trained with noisy labeled data from SD and unlabeled data from TD -- we name it wildly UDA (WUDA). We show that WUDA ruins all UDA methods if taking no care of label noise in SD, and to this end, we propose a Butterfly framework, a powerful and efficient solution to WUDA. Butterfly maintains four deep networks simultaneously, where two take care of all adaptations (i.e., noisy-to-clean, labeled-to-unlabeled, and SD-to-TD-distributional) and then the other two can focus on classification in TD. As a consequence, Butterfly possesses all the conceptually necessary components for solving WUDA. Experiments demonstrate that, under WUDA, Butterfly significantly outperforms existing baseline methods.