---
title: 'OSDA: Open Set Domain Adaptation'
url: https://www.emergentmind.com/topics/osda
type: topic
---

# OSDA: Open Set Domain Adaptation

Searching arXiv for recent and foundational OSDA papers to ground the article.
Open Set Domain Adaptation (OSDA) is the variant of unsupervised domain adaptation in which a labeled source domain and an unlabeled target domain do not share the same label space: the source contains only the known classes, whereas the target contains both shared classes and target-private unknown classes. The task is therefore dual: transfer recognition of shared categories across domain shift, and reject target samples that do not belong to any source class. This makes OSDA structurally different from closed-set UDA, because indiscriminate alignment of all target samples with the source induces negative transfer by pulling unknown-class target features into known-class source clusters [1804.10427, 2206.07551].

## 1. Formal problem setting

A standard formulation introduces a labeled source domain
\[
\mathcal{D}^s = \{(\mathbf{x}^s_i,y^s_i)\}_{i=1}^{n_s}
\]
with labels in a source class set \( \mathcal{Y}^s \), and an unlabeled target domain
\[
\mathcal{D}^t = \{\mathbf{x}^t_j\}_{j=1}^{n_t}
\]
whose full label space satisfies
\[
\mathcal{Y}^t = \mathcal{Y}^s \cup \mathcal{Y}^{unk},
\]
with \( \mathcal{Y}^{unk}\neq\emptyset \). The resulting classifier is expected to map target inputs into either one of the shared classes or a single reject label,
\[
f:\mathbb{R}^{d_0}\to\mathcal{Y}^s\cup\{\text{unk}\}.
\]
This formulation appears, with minor notational variation, across backpropagation-based OSDA, graph-based methods, and semantic-recovery variants [2110.12635, 1804.10427, 2105.02432].

Several papers also quantify the degree of label-space mismatch by the openness
\[
O = 1-\frac{|C_s|}{|C_t|},
\]
which increases as the proportion of target-private classes grows [2007.12360, 2303.05933]. In practice, OSDA must resolve two intertwined uncertainties: whether a target sample belongs to the shared label space at all, and, if it does, how to align it with the source under covariate shift.

A stricter variant is source-free OSDA (SF-OSDA), which removes access to source-domain samples during adaptation and assumes that only a pretrained source model and unlabeled target samples are available. This setting is motivated by privacy, confidentiality, or transmission constraints, and makes OSDA harder because adaptation must proceed without simultaneous source-target access [2312.03767, 2411.12558].

## 2. Core difficulties and evaluation criteria

The defining failure mode of OSDA is negative transfer. If a method matches source and target distributions as though the label spaces were identical, target samples from unknown classes are forced toward known-class regions, degrading both shared-class recognition and unknown rejection [1804.10427, 2206.07551]. This is why OSDA methods typically attempt some form of known/unknown separation before, during, or jointly with alignment.

The literature repeatedly identifies three technical difficulties. The first is covariate shift: \(p(x)\neq q(x)\), so a domain-invariant representation is required. The second is category shift: target-private classes exist but have no source counterparts. The third, emphasized in imbalanced OSDA, is label shift: \(p(y)\neq q(y)\), sometimes combined with intra-domain class imbalance, so a single global unknown-detection threshold can be brittle [2303.04393]. Source-free settings add a fourth difficulty: error accumulation from imperfect pseudo-labeling or imperfect known/unknown splitting, because no source samples remain available to stabilize training [2312.03767].

Evaluation in classification OSDA commonly uses the known-class average accuracy \(OS^*\), unknown accuracy \(UNK\), and a harmonic mean. One standard form is
\[
HOS=\frac{2\cdot OS^*\cdot UNK}{OS^*+UNK},
\]
while some papers write the same quantity as H-score [2110.12635, 2412.18105]. A recurrent criticism of the older \(OS\) metric is that it can under-weight unknown rejection when the number of known classes is large; ROS therefore argues for the harmonic mean because it penalizes imbalance between shared-class accuracy and unknown detection [2007.12360]. In semantic segmentation, the analogous H-score is computed from \(mIoU_{known}\) and \(IoU_{unk}\) [2405.19899, 2601.01439].

## 3. Main methodological families

OSDA methods can be grouped by how they separate unknowns from knowns and how they restrict alignment to the shared subspace.

| Family | Mechanism | Representative papers |
|---|---|---|
| Adversarial boundary learning | Classifier forms an unknown boundary; generator is driven either to align target samples with source-known regions or to reject them as unknown | [1804.10427], [2003.03787] |
| Unknown-aware adversarial alignment | Source and target-known are aligned while target-unknown is explicitly segregated | [2206.07551] |
| Progressive pseudo-labeling and rejection | Confident target-known samples are added gradually; low-confidence or rejected samples are treated separately | [2110.12635], [2202.06174], [2303.05933] |
| Teacher-student SF-OSDA | EMA teacher guides known/unknown splitting and stabilizes pseudo-labels | [2312.03767] |
| Unknown discovery or exploitation | Unknowns are clustered, synthesized, or used as hard negatives to reshape decision boundaries | [2203.03329], [2411.12558], [2412.18105] |
| Dual-space or multimodal separation | Unknowns are detected through routing-space inconsistency or prompt-gradient behavior | [2311.00285], [2505.13507] |

The earliest adversarial formulation, “Open Set Domain Adaptation by Backpropagation,” trains a classifier to place target samples on a boundary with target unknown probability \(t=0.5\), while the feature generator is adversarially encouraged to push each target either toward source-known regions or away as unknown [1804.10427]. “Mind the Gap” generalizes this line by decoupling separation and alignment into a Sample Separation Network and a Distribution Matching Network, with soft weights \(w_j\) used to exclude probable unknowns from alignment and with a mutual-learning module that transfers information between the two networks [2003.03787].

Unknown-aware adversarial learning further sharpens this distinction. UADAL introduces a three-way discriminator over source, target-known, and target-unknown, with an adversarial objective that aligns \(p_s\) and \(p_{tk}\) while driving \(p_{tu}\) away. Its theoretical analysis derives the optimal discriminator and rewrites the generator objective in terms of KL divergences, thereby formalizing the simultaneous alignment-and-segregation objective [2206.07551].

A second major family replaces global alignment with progressive or curriculum-style pseudo-labeling. OSLPP learns a common subspace by graph embedding, progressively selecting confident target-known samples and rejecting probable unknowns; selected and rejected samples then reshape the graph used in the next subspace update [2110.12635]. PGL decomposes the target hypothesis space into shared and unknown subspaces and progressively pseudo-labels the most confident known samples, while SF-PGL adapts this idea to the source-free case through balanced pseudo-labeling that selects confident samples from each category at a fixed ratio and uses class-specific confidence thresholds as uncertainty weights [2202.06174]. SPLOS similarly adopts self-paced learning, replacing manual threshold tuning with a self-tuned instructive threshold and a cross-domain mixup curriculum that moves from easier to harder target-known samples [2303.05933].

Recent source-free methods emphasize stable unknown separation. USD uses the Jensen–Shannon distance between a hard pseudo-label and a temporally ensembled teacher prediction, then fits a two-component Gaussian Mixture Model to separate target samples into known and unknown subsets. Its co-training objective combines instance-weighted pseudo-cross-entropy, information maximization, triplet regularization, temporal consistency, and curriculum guidance [2312.03767]. RRDA adopts a different two-stage design: first recall unknowns by generating synthetic target-domain features near the source classifier’s decision boundary, then refine the model with an extended classifier and any off-the-shelf SFDA objective such as SHOT or AaD [2411.12558].

A third trend rejects the assumption that all unknowns should remain a single amorphous reject class. SCDA explicitly argues that indiscriminately treating implicit target categories as one “unknown” can ignore decision boundaries among them; it therefore alternates between discovering target-private classes by clustering and restructuring the classifier to include newly discovered classes [2203.03329]. Closely related unknown-exploitation approaches mine high-confidence unknown samples and use them as hard negatives to tighten one-vs-all decision boundaries, either directly, after augmentation, or through GAN-generated synthetic negatives [2412.18105].

## 4. Benchmarks, protocols, and empirical patterns

Most classification papers evaluate on Office-31, Office-Home, and VisDA-C/VisDA. Office-31 commonly appears with 10 known and 21 unknown classes [1804.10427, 2110.12635], while Office-Home appears under more than one protocol: 25 known and 40 unknown in OSLPP, USD, and RRDA-style settings [2110.12635, 2312.03767, 2411.12558], but 20 known and 45 unknown in the OVANet-based unknown-exploitation study [2309.08964]. This suggests that direct cross-paper comparison requires attention to protocol, not only headline H-score.

Large domain gaps materially change behavior. “Mind the Gap” repurposes PACS for OSDA and reports that existing OSDA methods suffer a larger relative drop when the gap grows, with heavy overlap between known and unknown target features under large shifts [2003.03787]. Reproducibility is itself a documented issue: ROS reports that re-running public code for OSBP, STA, and UAN yields OS accuracies 1.3–4.9 percentage points lower than originally reported on the same splits [2007.12360].

Within a fixed protocol, benchmark progress has been substantial. OSLPP reports average HOS values of 87.4% on Office-31 and 67.0% on Office-Home [2110.12635]. UADAL reports approximately 88.1% and 68.7% HOS on the same datasets, with theoretical guarantees for unknown-aware alignment [2206.07551]. DSD, a threshold-free Mixture-of-Experts approach using inconsistencies between image-feature and routing-feature pseudo-labels, reports average HOS values of 94.2% on Office-31 and 75.5% on VisDA [2311.00285]. In source-free settings, USD reports 83.3% HOS on Office-31, 62.6% on Office-Home, and 69.4% on VisDA-C, while RRDA reports large improvements over naive open-set extensions of SHOT and AaD, including 93.8 HOS for SHOT+RRDA on Office-31 and 78.7 on VisDA [2312.03767, 2411.12558].

Alternative evaluation protocols extend beyond plain rejection. SR-OSDA introduces D2AwA and I2AwA to assess not only open-set recognition but also semantic attribute recovery for unseen categories, measuring seen accuracy \(S\), unseen accuracy \(U\), and harmonic mean \(H\) in a generalized zero-shot style protocol [2105.02432].

## 5. Extensions and generalizations

Several lines of work generalize classical OSDA rather than merely improving it.

Source-free OSDA removes source samples from adaptation. SF-PGL, USD, and RRDA all target this regime, but they differ in how they stabilize unknown separation: balanced pseudo-labeling in SF-PGL, JSD-plus-GMM teacher-student splitting in USD, and synthetic feature inversion with classifier expansion in RRDA [2202.06174, 2312.03767, 2411.12558]. A recurring theme is that source-free methods must explicitly learn a target-unknown feature space rather than merely exclude uncertain samples.

Imbalanced OSDA (IOSDA) augments the problem with intra-domain class imbalance and inter-domain label shift. OMEGA addresses this through moving-threshold estimation, unknown-aware target clustering, and gradual alignment via neighborhood clustering, reporting gains on Office-Home, VisDA-C, and DomainNet under imbalanced protocols [2303.04393]. This broadens OSDA from a covariate-shift problem with unknowns to a joint covariate-shift, label-shift, and category-mismatch problem.

Semantic enrichment is another extension. SCDA discovers multiple implicit target-private classes rather than treating all unknowns as a single output [2203.03329]. SR-OSDA goes further by attempting to recover semantic attributes for unseen target classes through structure-preserving partial alignment and attribute propagation over a visual graph [2105.02432]. A plausible implication is that OSDA can serve not only as a rejection mechanism but also as a discovery mechanism.

Open-set domain adaptation has also moved into dense prediction. BUS introduces OSDA for semantic segmentation (OSDA-SS), identifies boundary prediction and unknown-shape prediction as the central failures of closed-set UDA baselines, and addresses them with a dilation-erosion contrastive loss and OpenReMix augmentation [2405.19899]. SATS then argues explicitly for a two-stage framework—known/unknown separation followed by unknown-aware domain adaptation—reporting H-Score improvements of +3.85% on GTA5-to-Cityscapes and +18.64% on SYNTHIA-to-Cityscapes [2601.01439].

Multimodal and prompt-based variants have also appeared. A CLIP-based method introduces prompt-driven cross-domain alignment and gradient-aware open-set separation, using the \(L_2\)-norm of prompt gradients as a known/unknown detector on Office-Home [2505.13507]. Separately, the acronym OSDA has been reused in remote sensing for “Open-Set Discovery and Automatic Interpretation of Land-cover in Remote Sensing Imagery,” a three-stage framework combining SAM-style mask discovery, multimodal large language models, and LLM-as-judge evaluation [2509.18693].

## 6. Conceptual debates and open problems

One persistent misconception is that all unknowns can be collapsed into a single homogeneous reject class without loss. SCDA explicitly challenges this assumption, arguing that the decision boundaries between implicit categories are ignored when all unknowns are treated as one class [2203.03329]. DSD, RRDA, and semantic-recovery methods likewise introduce clustered or multi-subclass representations for unknowns, suggesting that “unknown” is often an operational label rather than a semantically coherent category [2311.00285, 2411.12558, 2105.02432].

A second unresolved issue is thresholding. Many OSDA pipelines still rely on confidence or entropy thresholds, whether fixed or heuristic. The literature contains several alternatives: self-tuned thresholds in SPLOS, Beta-mixture estimation in UADAL, two-component GMMs over JSD in USD, per-class moving thresholds in OMEGA, and a threshold-free dual-space mismatch rule in DSD [2303.05933, 2206.07551, 2312.03767, 2303.04393, 2311.00285]. This suggests that unknown detection remains as much a calibration problem as a representation problem.

Calibration and drift are recurrent concerns, especially in source-free settings. SF-PGL reports that balanced pseudo-labeling improves calibration and makes the trained model less prone to over-confident or under-confident predictions on target data [2202.06174]. USD attributes major gains to teacher-student co-training because student self-splitting leads to runaway error accumulation [2312.03767]. RRDA similarly emphasizes stable two-stage refinement after synthetic unknown recall [2411.12558].

Finally, the literature shows an active debate over one-stage versus two-stage training. Unified formulations attempt to discover unknowns and align knowns simultaneously, whereas SATS explicitly defends a separating-then-adapting strategy for semantic segmentation on the grounds that annotation imbalance between known and unknown classes induces negative transfer and underfitting in one-stage systems [2601.01439]. More broadly, open problems noted across the literature include automatic estimation of unknown-class proportion, scalability to many unknown classes, robustness under large domain gaps, extension beyond a single collapsed unknown head, and the move from mere rejection toward discovery, semantic attribution, and continual open-world adaptation [2110.12635, 2003.03787, 2405.19899].

Source: https://www.emergentmind.com/topics/osda