reweighted distribution alignment loss
AdversarialReweightingforPartial DomainAdaptation
Theconventional closed-set DAmethods generally assume that the source and target domains share the same label space. However, this assumption is often not realistic in practice.
Country:
- Asia > Middle East > Jordan (0.04)
- Asia > China (0.04)
Adversarial Reweighting for Partial Domain Adaptation Supplementary Material
The comparisons of the typical PDA methods are given in Table S-1. Mean Discrepancy (MMD) and the Jensen-Shannon (JS) divergence. ImageNet-Caltech, and VisDA-2017 are shown in Table S-4. This section illustrates the details for computing the Wasserstein distance discussed in Sect. With Eq. (S-7), the Wasserstein distance can be approximated by W (µ, ν) E Algorithm 1 presents the pseudo-code of our training algorithm in Sect.
Country:
- Asia > Middle East > Jordan (0.04)
- Asia > China > Shaanxi Province > Xi'an (0.04)
Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks (0.95)
Country:
- Asia > Middle East > Jordan (0.04)
- Asia > China > Shaanxi Province > Xi'an (0.04)
Technology: