Since DJP-MMD uses the source label and target pseudo label ˆYt∈Rnt to guide the model, incorrect pseudo-labeling might lead to the accumulation of errors. With this observation, Teng et al. introduce an adaptive adjustment strategy to progressively correct the mismatch during iteration [19]...
Therefore, this paper proposes the Joint Dynamic Maximum Mean Difference (JD-MMD) to deal with local subdomain adaptive as well as global adaptive. While minimizing the MMD, the LMMD is simultaneously used to introduce the target domain sample labels predicted by the network model into the ...