Deep domain adaptation based on multi-layer joint kernelized distance

Sitong Mao, Xiao Shen, Fu Lai Korris Chung

Research output: Chapter in book / Conference proceedingConference article published in proceeding or bookAcademic researchpeer-review

3 Citations (Scopus)

Abstract

Domain adaptation refers to the learning scenario where a model learned from the source data is applied on the target data which have the same categories but different distributions. In information retrieval, there exist application scenarios like cross domain recommendation characterized similarly. In this paper, by utilizing deep features extracted from the deep networks, we proposed to compute the multi-layer joint kernelized mean distance between the k th target data predicted as the i th category and all the source data of the j th category $d-ij k$. Then, target data $T-m$ that are most likely to belong to the i th category can be found by calculating the relative distance $d-ii k/\sum-j d-ij k$. By iteratively adding $T-m$ to the training data, the finetuned deep model can adapt on the target data progressively. Our results demonstrate that the proposed method can achieve a better performance compared to a number of state-of-the-art methods.

Original language English 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018 Association for Computing Machinery, Inc 1049-1052 4 9781450356572 https://doi.org/10.1145/3209978.3210155 Published - 27 Jun 2018 41st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018 - Ann Arbor, United StatesDuration: 8 Jul 2018 → 12 Jul 2018

Publication series

Name 41st International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018

Conference

Conference 41st Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2018 United States Ann Arbor 8/07/18 → 12/07/18