# Domain Adaptation without Source Data Source: [Domain Adaptation without Source Data]() Collection index: [llms.txt]() Data updated: 2025-10-25T23:30:26.199Z Authors: Youngeun Kim, Donghyeon Cho, Kyeongtak Han, Priyadarshini Panda, Sungeun Hong Published: 2020-07-03T07:21:30.000Z ## Abstract Domain adaptation assumes that samples from source and target domains are freely accessible during a training phase. However, such an assumption is rarely plausible in the real-world and possibly causes data-privacy issues, especially when the label of the source domain can be a sensitive attribute as an identifier. To avoid accessing source data that may contain sensitive information, we introduce Source data-Free Domain Adaptation (SFDA). Our key idea is to leverage a pre-trained model from the source domain and progressively update the target model in a self-learning manner. We observe that target samples with lower self-entropy measured by the pre-trained source model are more likely to be classified correctly. From this, we select the reliable samples with the self-entropy criterion and define these as class prototypes. We then assign pseudo labels for every target sample based on the similarity score with class prototypes. Furthermore, to reduce the uncertainty from the pseudo labeling process, we propose set-to-set distance-based filtering which does not require any tunable hyperparameters. Finally, we train the target model with the filtered pseudo labels with regularization from the pre-trained source model. Surprisingly, without direct usage of labeled source samples, our PrDA outperforms conventional domain adaptation methods on benchmark datasets. Our code is publicly available at https://github.com/youngryan1993/SFDA-SourceFreeDA ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Image Classification on Office-31 ResNet-50 (full)](): Image Classification; Office-31 ResNet-50 (full); Accuracy (A-\>D): 92.2. - [Image Classification on VisDA 2017 (test)](): Image Classification; VisDA 2017 (test); Class Accuracy (Plane): 86.9. - [Image Classification on Office-31](): Image Classification; Office-31; Average Accuracy: 87.2. - [Image Classification on Office-Home (test)](): Image Classification; Office-Home (test); Mean Accuracy: 65.7. ## Original sources and implementation [Original paper]() [PDF]()