Section: New Results
Improving Domain Adaptation By Source Selection
Domain adaptation consists in learning from a source data distribution a model that will be used on a different target data distribution. The domain adaptation procedure is usually unsuccessful if the source domain is too different from the target one. In [16], we study domain adaptation for image classification with deep learning in the context of multiple available source domains. This work proposes a multi-source domain adaptation method that selects and weights the sources based on inter-domain distances. We provide encouraging results on both classical benchmarks and a new real world application with 21 domains.