Timezone: »

 
Poster
Positive-Unlabeled Compression on the Cloud
Yixing Xu · Yunhe Wang · Hanting Chen · Kai Han · Chunjing XU · Dacheng Tao · Chang Xu

Tue Dec 10 10:45 AM -- 12:45 PM (PST) @ East Exhibition Hall B + C #117
in Deep Learning -- Efficient Inference Methods »

Many attempts have been done to extend the great success of convolutional neural networks (CNNs) achieved on high-end GPU servers to portable devices such as smart phones. Providing compression and acceleration service of deep learning models on the cloud is therefore of significance and is attractive for end users. However, existing network compression and acceleration approaches usually fine-tuning the svelte model by requesting the entire original training data (e.g. ImageNet), which could be more cumbersome than the network itself and cannot be easily uploaded to the cloud. In this paper, we present a novel positive-unlabeled (PU) setting for addressing this problem. In practice, only a small portion of the original training set is required as positive examples and more useful training examples can be obtained from the massive unlabeled data on the cloud through a PU classifier with an attention based multi-scale feature extractor. We further introduce a robust knowledge distillation (RKD) scheme to deal with the class imbalance problem of these newly augmented training examples. The superiority of the proposed method is verified through experiments conducted on the benchmark models and datasets. We can use only 8% of uniformly selected data from the ImageNet to obtain an efficient model with comparable performance to the baseline ResNet-34.

Keywords: Algorithms -> Semi-Supervised Learning · Deep Learning -> CNN Architectures

Author Information

Yixing Xu (Huawei Noah's Ark Lab)
Yunhe Wang (Huawei Noah's Ark Lab)
Hanting Chen (Peking University)
Kai Han (Huawei Noah's Ark Lab)
Chunjing XU (Huawei Technologies)
Dacheng Tao (University of Sydney)
Chang Xu (University of Sydney)

More from the Same Authors