6 papers
Feature Mining: A Novel Training Strategy for Convolutional Neural Network
Tianshu Xie, Xuan Cheng, Xiaomin Wang +3
In this paper, we propose a novel training strategy for convolutional neural network(CNN) named Feature Mining, that aims to strengthen the network's learning of the local feature.…
Go Small and Similar: A Simple Output Decay Brings Better Performance
Xuan Cheng, Tianshu Xie, Xiaomin Wang +3
Regularization and data augmentation methods have been widely used and become increasingly indispensable in deep learning training. Researchers who devote themselves to this have c…
Self-supervision of Feature Transformation for Further Improving Supervised Learning
Zilin Ding, Yuhang Yang, Xuan Cheng +2
Self-supervised learning, which benefits from automatically constructing labels through pre-designed pretext task, has recently been applied for strengthen supervised learning. Sin…
Self-supervised Feature Enhancement: Applying Internal Pretext Task to Supervised Learning
Yuhang Yang, Zilin Ding, Xuan Cheng +2
Traditional self-supervised learning requires CNNs using external pretext tasks (i.e., image- or video-based tasks) to encode high-level semantic visual representations. In this pa…
FocusedDropout for Convolutional Neural Network
Tianshu Xie, Minghui Liu, Jiali Deng +3
In convolutional neural network (CNN), dropout cannot work well because dropped information is not entirely obscured in convolutional layers where features are correlated spatially…
Cut-Thumbnail: A Novel Data Augmentation for Convolutional Neural Network
Tianshu Xie, Xuan Cheng, Minghui Liu +3
In this paper, we propose a novel data augmentation strategy named Cut-Thumbnail, that aims to improve the shape bias of the network. We reduce an image to a certain size and repla…