Skip to main content

Table 1 Domain adaptation model based on pseudo-labelling

From: Selecting pseudo supervision for unsupervised domain adaptive SAR target classification

Variable Algorithm
\(X_{s}\): Dataset in the source domain
\(X_{t}\): Dataset in the target domain
\(X_{l}\): Labelled training data set initialized as the source data \(X_{s}\)
\(G\): Classifier trained on \(T\)
\(Q\): Selective pseudo-labelling strategy
Initialization
Train the classifier \(G\) on the labelled training data set \(X_{l}\)
Assign pseudo-labels for all the data in \(X_{t}\) using the classifier \(G\)
Repeat:
Select pseudo-labelled samples in \(X_{t}\) according to \(Q\)
Add samples selected in (3) to \(X_{l}\)
Retrain the classifier \(G\)
Until the iteration condition is met or the pseudo-labels no longer change