A Kernelized Unified Framework for Domain Adaptation

Abstract

The performance of the supervised learning algorithms such as k-nearest neighbor (k-NN) depends on the labeled data. For some applications (Target Domain), obtaining such labeled data is very expensive and labor-intensive. In a real-world scenario, the possibility of some other related application (Source Domain) is always accompanied by sufficiently labeled data. However, there is a distribution discrepancy between the source domain and the target domain application data as the background of collecting both the domains data is different. Therefore, source domain application with sufficient labeled data cannot be directly utilized for training the target domain classifier. Domain Adaptation (DA) or Transfer learning (TL) provides a way to transfer knowledge from source domain application to target domain application. Existing DA methods may not perform well when there is a much discrepancy between the source and the target domain data, and the data is non-linear separable. Therefore, in this paper, we provide a Kernelized Unified Framework for Domain Adaptation (KUFDA) that minimizes the discrepancy between both the domains on linear or non-linear data-sets and aligns them both geometrically and statistically. The substantial experiments verify that the proposed framework outperforms state-of-the-art Domain Adaptation and the primitive methods (Non-Domain Adaptation) on real-world Office-Caltech and PIE Face data-sets. Our proposed approach (KUFDA) achieved mean accuracies of 86.83% and 74.42% for all possible tasks of Office-Caltech with VGG-Net features and PIE Face data-sets.

Unsupervised Transfer Learning via Relative Distance Comparisons

“…When searching for a new feature space for both domains, the searched space for the target domain may be irrelevant feature space [21]. Therefore, to ensure that the searched space is the relevant feature space, we need to maximize the target domain variance.…”

“…• ILS [39]: This method uses the Riemannian optimization techniques for matching the first and second-order statistics in the latent space of both domains. • KUFDA [21]: KUFDA improves JGSA by incorporating the original geometric structure of the data with a robust Laplacian term. • UnPSO [47]: By using the PSO algorithm, UnPSO selects a good subset of features across both domains and use MMD for minimizing the distribution gap between domains.…”

“…Geodesic Flow Kernel (GFK) [13] uses a set of intermediate subspaces to minimize the distribution gap. Joint Geometrical and Statistical Alignment (JGSA) [14] method follows the first four-objectives( from objectivei to objective-iv), but a Kernelized Unified Framework for Domain Adaptation (KUFDA) [21] method follows all of the objectives.…”

A Subspace Based Transfer Joint Matching with Laplacian Regularization for Visual Domain Adaptation

“…However, the feature space obtained by JGSA is not notable because data samples in this space may lose their original similarity so that they can be easily misclassified by the classifier. Kernelized Unified Framework for Domain Adaptation (KUFDA) [16] improves JGSA by adopting the original similarity weight matrix term so that the sample does not lose its original similarity in the learned space. KUFDA follows most of the above discussed important properties but still suffers from outlier data samples, and this is due to not considering instance re-weighting term.…”

“…Kernelized Unified Framework for Domain Adaptation (KUFDA) [16]: This TL method improves the JGSA method by adding the Laplacian regularization term.…”