Rademacher complexity bound for domain adaptation regression

Domain adaptation problems arise when the data distribution in test domain is different from that in training domain. In this paper, we provide a new error bound for the domain adaptive regression problem. Inspired by the original ideas in 0-1 classification, firstly, an error bound with a large amount of samples of source domain can be got in a new scene of regression. In the process, we utilize the covering number and Rademacher complexity respectively. Then we combine the error bound of source domain by Rademacher complexity with the divergence distance to get a new learning bound in regression. Using the thought and framework in classification to deal with error bound problems in regression is the key ideas, it opens the door to tackling domain adaptation tasks by making full use of the Rademacher complexity tools in the new scenario of regression.