基于类内最大均值差异的无监督领域自适应算法

Unsupervised domain adaptive algorithm with intra-class maximum mean discrepancy

传统的无监督领域自适应算法在对齐总体分布时存在分类信息流失问题,难以保证迁移学习效果。针对这个问题,提出了一种基于类内最大均值差异的分布对齐策略。该策略首先预测所有样本的伪标签,然后借助伪标签样本信息依次对齐每个类别的领域类内分布。在深度学习框架下,所提算法能够有效保留分类信息,提高了目标领域的预测能力。实验结果表明,与传统算法比较,所提算法在多个基准数据集上获得了最优的迁移学习效果。

In the unsupervised domain adaptation area,loss of label information is still an open problem during the alignment of the global distribution,and thus the effect of transfer learning is difficult to guarantee.To alleviate this problem,this paper proposed an algorithm which adopted a distribution alignment strategy based on the intra-class maximum mean discrepancy.This strategy firstly predicted the pseudo labels for all samples,then aligned the intra-class distributions of two domains with the help of the predicted labels.Under the deep learning framework,the proposed algorithm effectively avoided label information being washed away and greatly improved the prediction ability on the target domain.The experimental results show that the proposed algorithm outperforms the traditional algorithms on the benchmarks.