Deep Transfer Learning with Joint Adaptation Networks

Deep networks have been successfully applied to learn transferable features for adapting models from a source domain to a different target domain. In this paper, we present joint adaptation networks (JAN), which learn a transfer network by aligning the joint distributions of multiple domain-specific layers across domains based on a joint maximum mean discrepancy (JMMD) criterion. Adversarial training strategy is adopted to maximize JMMD such that the distributions of the source and target domains are made more distinguishable. Learning can be performed by stochastic gradient descent with the gradients computed by back-propagation in linear-time. Experiments testify that our model yields state of the art results on standard datasets.

PDF Abstract ICML 2017 PDF ICML 2017 Abstract


Introduced in the Paper:


Used in the Paper:

Office-Home HMDB51 Office-31 VisDA-2017
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Domain Adaptation HMDBfull-to-UCF JAN Accuracy 79.69 # 2
Multi-Source Unsupervised Domain Adaptation Office-Caltech10 JAN Accuracy 95.5 # 8
Unsupervised Domain Adaptation Office-Home JAN [cite:ICML17JAN] Accuracy 76.8 # 4
Domain Adaptation VisDA2017 JAN Accuracy 58.3 # 22

Results from Other Papers

Task Dataset Model Metric Name Metric Value Rank Source Paper Compare
Domain Adaptation UCF-to-HMDBfull JAN Accuracy 74.72 # 2


No methods listed for this paper. Add relevant methods here