Discriminative Joint Probability Maximum Mean Discrepancy (DJP-MMD) for Domain Adaptation

TL;DR

This paper introduces DJP-MMD, a novel metric for domain adaptation that improves distribution discrepancy measurement and enhances both transferability and class discriminability, leading to better classification results.

Contribution

It proposes DJP-MMD, a new discrepancy measure that simplifies and improves upon joint MMD, integrating transferability and discriminability in domain adaptation.

Findings

DJP-MMD outperforms traditional MMD on six image datasets.

The new metric provides a more accurate and simpler theoretical basis.

Embedding DJP-MMD improves classification performance.

Abstract

Maximum mean discrepancy (MMD) has been widely adopted in domain adaptation to measure the discrepancy between the source and target domain distributions. Many existing domain adaptation approaches are based on the joint MMD, which is computed as the (weighted) sum of the marginal distribution discrepancy and the conditional distribution discrepancy; however, a more natural metric may be their joint probability distribution discrepancy. Additionally, most metrics only aim to increase the transferability between domains, but ignores the discriminability between different classes, which may result in insufficient classification performance. To address these issues, discriminative joint probability MMD (DJP-MMD) is proposed in this paper to replace the frequently-used joint MMD in domain adaptation. It has two desirable properties: 1) it provides a new theoretical basis for computing the distribution discrepancy, which is simpler and more accurate; 2) it increases the transferability and discriminability simultaneously. We validate its performance by embedding it into a joint probability domain adaptation framework. Experiments on six image classification datasets demonstrated that the proposed DJP-MMD can outperform traditional MMDs.