Abstract
Domain adaptation is critical to transfer the invaluable source domain knowledge to the target domain. In this paper, for a particular visual attention model, saying hard attention, we consider to adapt the learned hard attention to the unlabeled target domain. To tackle this kind of hard attention adaptation, a novel adversarial reward strategy is proposed to train the policy of the target domain agent. In this adversarial training framework, the target domain agent competes with the discriminator which takes the attention features generated from the both domain agents as input and tries its best to distinguish them, and thus the target domain policy is learned to align the local attention feature to its source domain counterpart. We evaluated our model on the benchmarks of the cross-domain tasks, such as the centered digits datasets and the enlarged non-centered digits datasets. The experimental results show that our model outperforms the ADDA and other existing methods.
Highlights
In recent years, deep convolutional neural networks have achieved state-of-the-art performance in many visual tasks, such as image classification, object detection and semantic segmentation [1,2].in many practical cases, there exists problems of distribution mismatch [3] and domain shift [4] between different visual tasks, which results in poor generalization performance to the new task
We propose to train the target domain policy πT via a novel adversarial reward strategy in which πT is encouraged by positive reward if the collected hard attention feature is close enough to the source domain attention feature
Adapting attention from label-rich source domain to unlabeled target domain would inevitably improve the overall performance of transfer learning
Summary
Deep convolutional neural networks have achieved state-of-the-art performance in many visual tasks, such as image classification, object detection and semantic segmentation [1,2]. Since the emergence of the seminal work of generative adversarial networks (GAN) [6], adversarial adaptation methods have been attracting great attention [7,8,9,10,11], which explore the performance advantages of pitting two networks against each other to reduce the distribution difference between source and target domain. A hard attention model usually consists of a recurrent structure to select the most discriminative local features from image patches, and hard attention is non-differentiable due to sampling and cropping operations, which poses a challenge to adapt such hard attention from the source domain to the target domain. Different from the classical ADDA framework, to overcome the non-differentiable nature of hard attention, the most important contribution of this work is that we design an adversarial reward strategy to train the target agent T via reinforcement learning technique.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
