Dynamic Feature Alignment for Semi-supervised Domain Adaptation

Gongbo Liang; Nathan Jacobs; Yu Zhang

RECORD

Dynamic Feature Alignment for Semi-supervised Domain Adaptation

Yu Zhang, Gongbo Liang, Nathan Jacobs

Computer Vision and Pattern Recognition · Submitted 18 Oct 2021

abstract compiled · 2026-09-23 · no language-model review

The authors' main claim Our approach, which doesn't require extensive tuning or adversarial training, significantly improves the state of the art for semi-supervised domain adaptation.

how it is stated · asserted 1.00 read the record →

Comment (0) Read on arXiv ↗

Cite ▾

Claim paper

The compiled record

The paper's own sentences, arranged. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Open the compiled record

Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-09-23: 1 request, 5,551 input tokens.

What the abstract itself states, as probabilities: claims first 0.08 · states limitation 0.09 · human participants 0.03 · controlled comparison 0.34 · randomized assignment 0.02 · states quantitative result 0.06 · compares with prior results 0.82 · mentions code or data availability 0.02.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

Most research on domain adaptation has focused on the purely unsupervised setting, where no labeled examples in the target domain are available. However, in many real-world scenarios, a small amount of labeled target data is available and can be used to improve adaptation. We address this semi-supervised setting and propose to use dynamic feature alignment to address both inter- and intra-domain discrepancy. Unlike previous approaches, which attempt to align source and target features within a mini-batch, we propose to align the target features to a set of dynamically updated class prototypes, which we use both for minimizing divergence and pseudo-labeling. By updating based on class prototypes, we avoid problems that arise in previous approaches due to class imbalances. Our approach, which doesn't require extensive tuning or adversarial training, significantly improves the state of the art for semi-supervised domain adaptation. We provide a quantitative evaluation on two standard datasets, DomainNet and Office-Home, and performance analysis.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	Most research on domain adaptation has focused on the purely unsupervised setting, where no labeled examples in the target domain are available.	background 1.00	0.04	0.02	asserted 1.00
s1	However, in many real-world scenarios, a small amount of labeled target data is available and can be used to improve adaptation.	background 0.79	0.06	0.03	asserted 0.84
s2	We address this semi-supervised setting and propose to use dynamic feature alignment to address both inter- and intra-domain discrepancy.	contribution 0.89	0.83	0.11	asserted 0.92
s3	Unlike previous approaches, which attempt to align source and target features within a mini-batch, we propose to align the target features to a set of dynamically updated class prototypes, which we use both for minimizing divergence and pseudo-labeling.	method 0.84	0.82	0.08	asserted 0.99
s4	By updating based on class prototypes, we avoid problems that arise in previous approaches due to class imbalances.	method 0.47	0.63	0.06	asserted 1.00
s5	Our approach, which doesn't require extensive tuning or adversarial training, significantly improves the state of the art for semi-supervised domain adaptation.	result 0.88	0.86	0.82	asserted 1.00
s6	We provide a quantitative evaluation on two standard datasets, DomainNet and Office-Home, and performance analysis.	contribution 0.68	0.94	0.05	asserted 0.78

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.93.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

The paper

Original abstract

Most research on domain adaptation has focused on the purely unsupervised setting, where no labeled examples in the target domain are available. However, in many real-world scenarios, a small amount of labeled target data is available and can be used to improve adaptation. We address this semi-supervised setting and propose to use dynamic feature alignment to address both inter- and intra-domain discrepancy. Unlike previous approaches, which attempt to align source and target features within a mini-batch, we propose to align the target features to a set of dynamically updated class prototypes, which we use both for minimizing divergence and pseudo-labeling. By updating based on class prototypes, we avoid problems that arise in previous approaches due to class imbalances. Our approach, which doesn't require extensive tuning or adversarial training, significantly improves the state of the art for semi-supervised domain adaptation. We provide a quantitative evaluation on two standard datasets, DomainNet and Office-Home, and performance analysis.

Human review

How Pith checks papers → · Full audit of this paper →

How the literature uses it

What this paper does with the arXiv papers it cites

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

What it cites · 1 sentence without a confident label

Uses the method or tool of 1 paper

We implement our model using PyTorch [ref] .

PyTorch: An Imperative Style, High-Performance Deep Learning Library · Implementation Details · 1.00

Compares with 2 papers

Baselines include training a network only using labeled samples (S+T), a entropy minimization based semi-supervised method (ENT [ref] ), three feature alignment-based unsupervised domain adaptation models (DANN [ref] , ADR [ref] , and CDAN [ref] ), and three state-of-the-art semi-supervised learning models (SagNet [ref] , MME [ref] , and APE [ref] ) that aim to the same goal as our method.

Conditional Adversarial Domain Adaptation · Baselines · 0.97
Baselines include training a network only using labeled samples (S+T), a entropy minimization based semi-supervised method (ENT [ref] ), three feature alignment-based unsupervised domain adaptation models (DANN [ref] , ADR [ref] , and CDAN [ref] ), and three state-of-the-art semi-supervised learning models (SagNet [ref] , MME [ref] , and APE [ref] ) that aim to the same goal as our method.

Adversarial Dropout Regularization · Baselines · 0.97

Cites as background 1 paper

This problem is addressed as semi-supervised domain adaptation (SSDA) [ref] .

Few-Shot Adversarial Domain Adaptation · Semi-Supervised Domain Adaptation · 0.99

Discussion

Know something about this paper? Leave a comment, or highlight any passage above to comment on it.