# Continual Source-Free Unsupervised Domain Adaptation

arXiv: https://arxiv.org/abs/2304.07374 · HTML: https://mlsanity.com/paper/2304.07374/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 32 references that are in the map.
- First posted on arXiv 14 Apr 2023.

## Builds on (most cited first, 24 of 24)

- [Moment Matching for Multi-Source Domain Adaptation](https://mlsanity.com/paper/1812.01754/) (2018)
- [Deeper, Broader and Artier Domain Generalization](https://mlsanity.com/paper/1710.03077/) (2017)
- [Do We Really Need to Access the Source Data? Source Hypothesis Transfer for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2002.08546/) (2020)
- [Generalizing to Unseen Domains via Adversarial Data Augmentation](https://mlsanity.com/paper/1805.12018/) (2018)
- [Domain Generalization by Solving Jigsaw Puzzles](https://mlsanity.com/paper/1903.06864/) (2019)
- [Dreaming to Distill: Data-free Knowledge Transfer via DeepInversion](https://mlsanity.com/paper/1912.08795/) (2019)
- [Understanding Deep Image Representations by Inverting Them](https://mlsanity.com/paper/1412.0035/) (2014)
- [Rainbow Memory: Continual Learning with a Memory of Diverse Samples](https://mlsanity.com/paper/2103.17230/) (2021)
- [Exploiting Deep Generative Prior for Versatile Image Restoration and Manipulation](https://mlsanity.com/paper/2003.13659/) (2020)
- [Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification](https://mlsanity.com/paper/2001.01526/) (2020)
- [Progressive Feature Alignment for Unsupervised Domain Adaptation](https://mlsanity.com/paper/1811.08585/) (2018)
- [Generalized Source-free Domain Adaptation](https://mlsanity.com/paper/2108.01614/) (2021)
- [Domain Adaptation without Source Data](https://mlsanity.com/paper/2007.01524/) (2020)
- [Discriminative Adversarial Domain Adaptation](https://mlsanity.com/paper/1911.12036/) (2019)
- [Online Meta-Learning for Multi-Source and Semi-Supervised Domain Adaptation](https://mlsanity.com/paper/2004.04398/) (2020)
- [Unsupervised Multi-Target Domain Adaptation: An Information Theoretic Approach](https://mlsanity.com/paper/1810.11547/) (2018)
- [Self-supervised Domain Adaptation for Computer Vision Tasks](https://mlsanity.com/paper/1907.10915/) (2019)
- [Curriculum Manager for Source Selection in Multi-Source Domain Adaptation](https://mlsanity.com/paper/2007.01261/) (2020)
- [Continual Adaptation of Visual Representations via Domain Randomization and Meta-learning](https://mlsanity.com/paper/2012.04324/) (2020)
- [Empirical Bayes Transductive Meta-Learning with Synthetic Gradients](https://mlsanity.com/paper/2004.12696/) (2020)
- [Continual Learning via Bit-Level Information Preserving](https://mlsanity.com/paper/2105.04444/) (2021)
- [Dynamic Transfer for Multi-Source Domain Adaptation](https://mlsanity.com/paper/2103.10583/) (2021)
- [Generative Pseudo-label Refinement for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2001.02950/) (2020)
- [Knowledge Distillation Methods for Efficient Unsupervised Adaptation Across Multiple Domains](https://mlsanity.com/paper/2101.07308/) (2021)

## Built on by (most cited first, 1 of 1)

- [A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learning](https://mlsanity.com/paper/2307.09218/) (2023)

## Abstract (arXiv)

Existing Source-free Unsupervised Domain Adaptation (SUDA) approaches inherently exhibit catastrophic forgetting. Typically, models trained on a labeled source domain and adapted to unlabeled target data improve performance on the target while dropping performance on the source, which is not available during adaptation. In this study, our goal is to cope with the challenging problem of SUDA in a continual learning setting, i.e., adapting to the target(s) with varying distributional shifts while maintaining performance on the source. The proposed framework consists of two main stages: i) a SUDA model yielding cleaner target labels -- favoring good performance on target, and ii) a novel method for synthesizing class-conditioned source-style images by leveraging only the source model and pseudo-labeled target data as a prior. An extensive pool of experiments on major benchmarks, e.g., PACS, Visda-C, and DomainNet demonstrates that the proposed Continual SUDA (C-SUDA) framework enables preserving satisfactory performance on the source domain without exploiting the source data at all.