Source-free and black-box domain adaptation via distributionally adversarial training *

Source-free unsupervised domain adaptation is one class of practical deep learning methods which gen-eralize in the target domain without transferring data from source domain. However, existing source-free domain adaptation methods rely on source model transferring. In many data-critical scenarios, the trans-ferred source models may suffer from membership inference attacks and expose private data. In this paper, we aim to overcome a more practical and challenging setting where the source models cannot be transferred to the target domain. The source models are considered as queryable black-box models which only output hard labels. We use public third-party data to probe the source model and obtain super-vision information, dispensing with transferring source model. To fill the gap between third-party data and target data, we further propose Distributionally Adversarial Training (DAT) to align the distribution of third-party data with target data, gain more informative query results and improve the data efficiency. We call this new framework Black-box Probe Domain Adaptation (BPDA) which adopts query mechanism and DAT to probe and refine supervision information. Experimental results on several domain adapta-tion datasets demonstrate the practicability and data efficiency of BPDA in query-only and source-free unsupervised domain adaptation.& COPY; 2023 Elsevier Ltd. All rights reserved.