Journal
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
Volume 41, Issue 12, Pages 3071-3085Publisher
IEEE COMPUTER SOC
DOI: 10.1109/TPAMI.2018.2868685
Keywords
Task analysis; Kernel; Adaptation models; Convolutional neural networks; Gallium nitride; Testing; Domain adaptation; deep learning; convolutional neural network; two-sample test; multiple kernel learning
Funding
- National Key R&D Program of China [2016YFB1000701]
- National Natural Science Foundation of China [61772299, 71690231, 61502265]
- DARPA Program on Lifelong Learning Machines
Ask authors/readers for more resources
Domain adaptation studies learning algorithms that generalize across source domains and target domains that exhibit different distributions. Recent studies reveal that deep neural networks can learn transferable features that generalize well to similar novel tasks. However, as deep features eventually transition from general to specific along the network, feature transferability drops significantly in higher task-specific layers with increasing domain discrepancy. To formally reduce the effects of this discrepancy and enhance feature transferability in task-specific layers, we develop a novel framework for deep adaptation networks that extends deep convolutional neural networks to domain adaptation problems. The framework embeds the deep features of all task-specific layers into reproducing kernel Hilbert spaces (RKHSs) and optimally matches different domain distributions. The deep features are made more transferable by exploiting low-density separation of target-unlabeled data in very deep architectures, while the domain discrepancy is further reduced via the use of multiple kernel learning that enhances the statistical power of kernel embedding matching. The overall framework is cast in a minimax game setting. Extensive empirical evidence shows that the proposed networks yield state-of-the-art results on standard visual domain-adaptation benchmarks.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available