Positive-Unlabeled Compression on the Cloud

TL;DR

This paper introduces a positive-unlabeled learning approach combined with knowledge distillation to efficiently compress and accelerate CNN models on the cloud using minimal labeled data and abundant unlabeled data.

Contribution

It proposes a novel PU-based training framework with attention-based feature extraction and robust knowledge distillation for cloud-based model compression.

Findings

Achieves comparable performance with only 8% of ImageNet data

Effectively handles class imbalance in augmented training data

Demonstrates superior results on benchmark models and datasets

Abstract

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 (\eg 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.