---
title: Federated Source-Free Domain Adaptation
url: https://www.emergentmind.com/topics/federated-source-free-domain-adaptation-ffreeda-996f4da8-c324-406b-b74b-d6585d7ff6ae
type: topic
---

# Federated Source-Free Domain Adaptation

Searching arXiv for recent papers on Federated Source-Free Domain Adaptation to ground the article in the cited literature.
Federated Source-Free Domain Adaptation, written in the literature as **FFREEDA** or **FFreeDA**, denotes a federated learning setting in which a server first trains a model on labeled source-domain data, then adapts that model collaboratively across multiple clients that hold only **unlabeled** target-domain data, while the **source dataset is unavailable during adaptation**. The setting combines federated learning, source-free domain adaptation, and unsupervised client-side training, and it has been studied in both **semantic segmentation** and **classification**. Its central difficulty is that the system must transfer a source-pretrained model across heterogeneous target clients without revisiting source samples and without access to target labels, so pseudo-label quality, client drift, and cross-client domain heterogeneity become first-order concerns [2210.02326; 2412.13757; 2512.06738].

## 1. Problem formulation and scope

A standard FFREEDA formulation begins with a labeled source dataset on the server and unlabeled target datasets on clients. One classification-oriented definition writes the source dataset as
\[
D_s = \{(x_s^i, y_s^i)\ |\ i=1,\ldots,N_s\},
\]
and the target data on each of \(K\) clients as
\[
D_t^{(k)} = \{x_t^{(k),i}\ |\ i=1,\ldots,N_t^{(k)}\}.
\]
The workflow is: the server trains a source model using labeled source data, broadcasts that pretrained source model to clients, each client adapts the model using its unlabeled target-domain data, and the server aggregates client updates to obtain one global model adapted to the target domain. In the same literature, the source model is often decomposed as \(f^s = h^s \circ g^s\), with a feature extractor \(g^s\) and classifier \(h^s\), and the client distributions are explicitly heterogeneous, with
\[
\mathbb{P}_{k'}(x) \neq \mathbb{P}_{l'}(x)
\]
for different clients or domains [2509.08372; 2512.06738].

In semantic segmentation, the task is framed analogously but with pixel-wise outputs and client-side unlabeled image collections. One formal objective is
\[
w^* = \arg\min_w \sum_{k\in[K]} \frac{|\mathcal{D}_k^T|}{|\mathcal{D}^T|}L_k(w),
\]
where \(\mathcal{D}^T=\bigcup_{k \in K} \mathcal{D}^T_k\). In that formulation, the source and target share the same semantic label set, but target clients are non-iid and may be drawn from latent visual domains [2210.02326].

The scope of the setting has expanded. Early work introduced FFREEDA for **semantic segmentation** under server-only source pre-training and unlabeled clients. Later work extended FFREEDA to **classification**, emphasizing that prior FFREEDA work had focused only on semantic segmentation. Subsequent studies further specialized the setting to **Class Imbalanced FFREEDA (CI-FFREEDA)**, where class imbalance appears in the source domain, in target domains, and across target clients, and to multi-viewpoint semantic segmentation with **car** and **drone** clients under **adverse weather** [2210.02326; 2412.13757; 2509.08372; 2403.13762].

## 2. Core technical difficulties

FFREEDA is difficult because no labeled source data are available during adaptation, clients are unlabeled, and large domain gaps can exist both between source and target and across clients. In source-free adaptation, clients therefore rely on pseudo-labels; under large domain shifts, these pseudo-labels become unreliable, and in federated training poor pseudo-labels cause **client drift**, where local models overfit their own biased pseudo-label distributions and aggregated updates become suboptimal. This challenge is explicit in classification work that identifies the core problem as adapting a pretrained server model to multiple heterogeneous unlabeled client domains, without source data, while preventing pseudo-label collapse and client drift [2512.06738].

A second difficulty is that heterogeneity is not limited to covariate shift. CI-FFREEDA introduces **intra-domain class imbalance**, **inter-domain label shift**, and **inter-client label shift**, with relations such as
\[
p(y_s) \neq q(y_t),
\]
and, in the federated case,
\[
p(x_s) \neq q(x_t^{1}) \neq \ldots \neq q(x_t^{K}), \qquad
p(y_s) \neq q(y_t^{1}) \neq \ldots \neq q(y_t^{K}).
\]
That extension is presented as more realistic than standard FFREEDA because it combines source imbalance, target imbalance, source-target label shift, client-to-client label shift, and domain gap [2509.08372].

A third difficulty arises in semantic segmentation when heterogeneity reflects not only domain appearance but also sensing geometry and environmental conditions. The adverse-weather, multi-viewpoint formulation includes both car and drone clients, with client populations that are unlabeled, non-i.i.d., unbalanced, and semantically or viewpoint heterogeneous. The target side is partitioned across clients with disjoint local datasets, while weather varies dynamically during training. This setting compounds standard source-free self-training instability with viewpoint shift and weather shift [2403.13762].

These formulations collectively establish that FFREEDA is not merely federated learning with missing labels, nor merely source-free domain adaptation distributed across devices. It is the joint problem of privacy-preserving adaptation, unlabeled multi-client optimization, and severe statistical heterogeneity [2210.02326].

## 3. Dominant methodological patterns

A common baseline pattern is **local adaptation with pseudo-labeling** followed by federated aggregation. In one classification formulation, target pseudo-labels are generated from class prototypes,
\[
c_{j} = \frac {\sum_{x_t\in D^t} \delta _j({f^t}(x_t))\:  {g^t}(x_t)}{\sum_{x_t\in D^t}\delta _j( {f^t}(x_t))},
\qquad
\hat {y}^i_t = \arg \min _j D_f({g^t}(x^i_t), c_j),
\]
and the local loss combines Information Maximization and pseudo-label cross-entropy. In its federated version, each client independently optimizes that loss and the server aggregates via FedAvg. This baseline is termed **FedLoA**, and it is described as weak in FFREEDA because each client self-trains on noisy pseudo-labels, pseudo-label noise differs across domains, local optimization drifts toward client-specific minima, and aggregating drifting client models harms the global model [2512.06738].

A second methodological pattern is **clustering and personalization**. In semantic segmentation, **LADD** combines source pre-training enhanced by style transfer, local self-training with regularization, and a federated clustered aggregation scheme based on clients’ style. The model is split into global parameters \(\phi\) and cluster-specific parameters \(\theta_c\), with global aggregation for \(\phi\) and cluster-specific aggregation for \(\theta_c\). In classification, **FedWCA** replaces style exchange with private, parameter-free clustering using model parameters, specifically the first-layer parameters of each client’s feature extractor, and then performs weighted aggregation across cluster models rather than strict within-cluster averaging [2210.02326; 2412.13757].

A third pattern is **consistency and alignment regularization**. **FedSCAl** keeps a local adaptation backbone but adds **Server-Client Alignment (SCAl)**, a regularizer that aligns client and server predictions. SCAl has two terms: a client alignment loss that matches weakly and strongly augmented client predictions under a confidence threshold, and a server alignment loss that matches client predictions to the server model’s confident predictions. The full local client objective is
\[
l_k(\mathbf{w}) = L_k^{LoA} + L_k^{SCAl},
\qquad
L_k^{SCAl} = \lambda_l L_k^l + \lambda_g L_k^g.
\]
This design uses the server’s pretrained model as a stabilizing teacher signal during federated adaptation, without source data [2512.06738].

A fourth pattern is **representation-first design**. The CI-FFREEDA study argues that many existing methods are improving the wrong part of the pipeline and proposes replacing the conventional CNN backbone with a **frozen vision foundation model**, specifically **DINOv2**, while training and aggregating only the bottleneck and classifier. The paper’s central claim is that the backbone matters more than the choice of FFREEDA algorithm, and it uses existing SFDA methods such as SHOT, NRC, AaD, ISFDA, and ICPR largely as heads on top of a stronger frozen representation [2509.08372].

| Paper | Task | Main mechanism |
|---|---|---|
| LADD [2210.02326] | Semantic segmentation | Style-driven clustered aggregation, KD, SWA teacher |
| FedWCA [2412.13757] | Classification | Parameter-free clustering, weighted cluster aggregation, improved pseudo-labeling |
| HyperFLAW [2403.13762] | Semantic segmentation | Weather-aware BN, hyperbolic prototypes, server smoothing |
| FedSCAl [2512.06738] | Classification | Server-client alignment with adaptive thresholding |
| CI-FFREEDA VFM study [2509.08372] | Classification under class imbalance | Frozen DINOv2 backbone, balanced sampling |

## 4. Representative algorithms

**LADD** introduced FFREEDA as a more realistic semantic-segmentation setting in which dense pixel annotations are unavailable on clients and the server can use labeled source data only during pre-training. Its architecture uses FDA-style Fourier amplitude statistics as compact style representations, clusters clients according to style vectors by selecting the best clustering with silhouette score, and maintains one global parameter set plus cluster-specific layers. Local training combines pseudo-label self-training, knowledge distillation from the pre-trained model, and an SWA-style teacher averaging mechanism:
\[
\mathcal{L} = \mathcal{L}_{pseudo} + \lambda_{kd}\mathcal{L}_{kd}.
\]
At test time, the style of a test image is extracted, compared to cluster centroids by L2 distance, and the corresponding cluster model is used [2210.02326].

**FedWCA** extends FFREEDA to classification and defines the goal as learning personalized models \(h_k = g_k \circ f_k\) for each client. Its first phase clusters clients using only model parameters and the **FINCH** procedure. Its second phase builds a weighted mixture of cluster models through client-specific weights \(\alpha_k\) and pseudo-performance weights \(\beta_k\), producing an initial model
\[
f_k^{\rm init} = \sum_c v_{k,c} f_c.
\]
Its third phase performs source-free local adaptation using a SHOT-style objective, prototype-based pseudo-labeling, fixed pseudo-labels within each communication round, two-model pseudo-label selection, and a mixup step for mismatched pseudo-labels with \(\mu=0.55\) in experiments [2412.13757].

**FedSCAl** addresses the observation that simple federated adaptation of source-free methods struggles with client drift in real-world scenarios due to extreme data heterogeneity. At each communication round, the server sends current global weights to selected clients, clients train locally for \(E\) epochs, return updated parameters, and the server aggregates them. The distinguishing component is SCAl. For client alignment, weak-augmented predictions are converted to one-hot pseudo-labels and regularize strong-augmented predictions via KL divergence under a confidence threshold \(\tau\). For server alignment, the one-hot prediction of the server model on weak augmentation supervises the client’s strong-augmentation prediction, again gated by confidence. The method also proposes **adaptive thresholding** based on prediction entropy, using
\[
H(\delta(f_k(x^i_k))) = - \sum_{j} \delta_j(f_k(x^i_k)) \log \left( \delta_j(f_k(x^i_k)) \right),
\]
and
\[
\gamma = 3*\frac{\mu_H - M_H}{\sigma_H},
\qquad
\tau = \tau_{init} + \max(\gamma_L,\min(\gamma,\gamma_H)).
\]
The stated interpretation is that positive skew indicates more confident predictions and tightens the threshold, whereas negative skew relaxes it [2512.06738].

**HyperFLAW** specializes FFREEDA to semantic segmentation with car and drone clients under clear day, night, rain, and fog. The method combines pseudo-label self-training with prototype clustering,
\[
\mathcal{L}_k(\theta)= \mathcal{L}_{st}(\theta) + \lambda_{cl}\mathcal{L}_{cl}(\theta),
\]
a weather-aware batch-normalization strategy that keeps separate BN parameters for each weather condition, and prototype alignment in hyperbolic space using the Poincaré ball model. Curvature is treated as learnable and updated on clients with Riemannian Adam, while global model aggregation is stabilized by averaging the current round’s federated model with a queue of previous global models [2403.13762].

The **CI-FFREEDA VFM** study departs from loss-centric and aggregation-centric design. It keeps the standard model form \(f(x)=h(g(x))\), replaces \(g\) with a frozen DINOv2 backbone, freezes backbone parameters during both source training and target adaptation, trains only the bottleneck and classifier, and uses balanced sampling in source training. It explicitly notes that, unlike some prior SFDA methods that freeze the classifier, the classifier remains trainable because a fixed backbone can reduce bottleneck expressiveness, especially when imbalance ratios vary across clients [2509.08372].

## 5. Experimental regimes and reported performance

The empirical literature spans both segmentation and classification, with differing metrics and benchmark conventions. Semantic-segmentation papers report **mIoU**, while classification papers report **classification accuracy (%)** or **macro averaged recall (MAR)**. Evaluation settings include GTA5 as source with Cityscapes, CrossCity, and Mapillary as targets for segmentation; Office-Home, DomainNet, Office-31, Digit-Five, PACS, and VisDA for classification; and additional adverse-weather evaluation on ACDC and FLYAWARE-R [2210.02326; 2403.13762; 2412.13757; 2509.08372; 2512.06738].

For semantic segmentation, LADD reported the following representative results. On **GTA5 \(\rightarrow\) Cityscapes**, Source only achieved \(24.05 \pm 1.14\), FedAvg + Self-Tr. \(35.10 \pm 0.73\), and LADD \(36.49 \pm 0.13\) or \(36.49 \pm 0.14\) depending on the variant. On **GTA5 \(\rightarrow\) CrossCity**, Source only achieved \(26.49 \pm 1.46\), MCD federated \(24.80 \pm 1.56\), FedAvg + Self-Tr. \(33.59 \pm 1.25\), and LADD \(39.87 \pm 0.14\) or \(40.09 \pm 0.19\). On **GTA5 \(\rightarrow\) Mapillary**, Source only achieved \(32.40 \pm 0.71\), FedAvg + Self-Tr. \(38.97 \pm 0.21\), and LADD \(40.16 \pm 1.02\) or \(38.78 \pm 1.82\). The strongest gains were reported on the more heterogeneous CrossCity and Mapillary setups [2210.02326].

HyperFLAW evaluated a harder benchmark with car and drone clients in adverse weather. On the main benchmark, Source Only achieved **22.70 mIoU**, FedAvg+ST **26.10**, LADD **26.96**, and HyperFLAW **28.96**. On the larger ACDC+FA-R-XL setup, HyperFLAW again outperformed the baselines, reaching **28.84 mIoU** versus **26.86** for LADD and **24.59** for FedAvg+ST. The same paper introduced **FLYAWARE-S** with **108,180 generated images** and **FLYAWARE-R** with **1,287 domain-translated images** [2403.13762].

For classification, FedWCA reported mean accuracy across clients and target domains over 15 trials. Its average accuracies were **77.86\%** on Digit-Five, **78.38\%** on PACS, and **65.97\%** on Office-Home, outperforming Source Only, Local, FedAvg, LADD, FedPCL+PL, and extended FL baselines such as FedProx, FedAMP, and IFCA [2412.13757].

FedSCAl evaluated a cross-silo setup with an **ImageNet-pretrained ViT-Small** backbone on Office-Home, DomainNet, and Office-31. On Office-Home, examples include Source = Art with FedLoA average **75.55** versus FedSCAl **79.05**, Source = Clipart with **83.18** versus **86.01**, Source = Product with **71.94** versus **75.56**, and Source = Real-World with **73.08** versus **76.59**. The reported gains over FedLoA were **+3.50\%** for Art, **+3.51\%** for Real-World, **+2.83\%** for Clipart, and **+3.62\%** for Product. On DomainNet, examples include Quickdraw-pretrained with FedLoA **48.24** and FedSCAl **60.61**, and Real-pretrained with **70.15** and **78.08**. On Office-31, Amazon-pretrained yielded **96.15** for FedLoA and **96.46** for FedSCAl, DSLR-pretrained **88.39** and **89.25**, and Webcam-pretrained **88.56** and **90.76**. An ablation on Office-Home reported averages of **75.93** with neither alignment term, **77.46** with client alignment only, **77.66** with server alignment only, and **79.30** with both [2512.06738].

The CI-FFREEDA VFM study reported that replacing a ResNet backbone with frozen DINOv2 features substantially changed outcomes. On Office-Home, averaged over source-target pairs, **ResNet-50** achieved **63.2 MAR**, **DINOv2 ViT-S** **73.7 MAR**, and **DINOv2 ViT-B** **79.8 MAR**, corresponding to gains of **+10.5** and **+16.6**. On VisDA, averaged over methods, **ResNet-101** achieved **73.2**, while **ViT-S / ViT-B** achieved **77.4**, a gain of **+4.2**. The same study also quantified robustness to domain gaps and imbalance: in domain adaptation without class imbalance, the source-to-target drop was **-17.6** for ResNet-50, **-10.6** for ViT-S, and **-7.3** for ViT-B; under inter-domain label shift, the drop from source performance to imbalanced target performance was **-21.8**, **-16.0**, and **-11.8**, respectively [2509.08372].

## 6. Limitations, contested interpretations, and practical implications

A persistent disagreement in the FFREEDA literature concerns the primary bottleneck. FedSCAl, FedWCA, LADD, and HyperFLAW each improve performance through better aggregation, clustering, alignment, or stabilization mechanisms. By contrast, the CI-FFREEDA work argues that existing SFDA methods and FL aggregation tricks help less than expected under real-world imbalance, and that the **feature extractor is the main lever**. This does not invalidate algorithmic advances; rather, it establishes a live research tension between **representation strength** and **adaptation or federation design** [2509.08372; 2512.06738; 2412.13757].

Several limitations are explicit. FedSCAl assumes a **predefined number of classes** across clients, which may reduce applicability when class sets evolve or are unknown, and it still depends on a reasonably useful source-pretrained server model, confidence thresholding, and pseudo-label quality. HyperFLAW remains challenged by severe conditions such as night, relies on a weather classifier plus curated weather-specific BN branches, and points toward multimodal data and more advanced weather adaptation strategies as future work. LADD’s clustering depends on exchanged style representations, whereas FedWCA presents its parameter-based clustering partly as a response to the privacy concerns raised by style feature sharing [2512.06738; 2403.13762; 2210.02326; 2412.13757].

Practical considerations are equally important. FedSCAl reports only a modest extra computation cost due to additional alignment forward passes and unchanged communication cost because only model parameters are exchanged. The frozen-VFM approach pushes this further: because the backbone is frozen, only the bottleneck and classifier are updated and transmitted, and those together are **less than 1 MB**, compared with roughly **94 MB** for ResNet-50 and **169 MB** for ResNet-101. The same study reports **11.0 G FLOPs** for frozen ViT-S, **43.9 G FLOPs** for frozen ViT-B, and, with feature banking, **0.61 M FLOPs** for ViT-S and **1.20 M FLOPs** for ViT-B. This suggests that some FFREEDA deployments may become more practical not by increasing adaptation complexity but by freezing a stronger backbone [2512.06738; 2509.08372].

Across these variants, a common conclusion emerges: FFREEDA is a realistic benchmark for privacy-preserving adaptation because the source domain is unavailable during target training, client data are unlabeled, and target distributions are heterogeneous. The literature diverges on whether the most effective remedy is clustered personalization, alignment-based regularization, weather- and geometry-aware adaptation, or stronger pretrained representations. A plausible implication is that future FFREEDA systems will increasingly combine these strands rather than treat them as mutually exclusive.

Source: https://www.emergentmind.com/topics/federated-source-free-domain-adaptation-ffreeda-996f4da8-c324-406b-b74b-d6585d7ff6ae