This paper proposes a new confidential deep learning (DL) inference system with ARM TrustZone to provide confidentiality and integrity of DL models and data in an untrusted edge device with limited memory. Although ARM TrustZone supplies a strong, hardware-supported trusted execution environment for protecting sensitive code and data in an edge device against adversaries, resource limitations in typical edge devices have raised significant challenges for protecting on-device DL requiring large memory consumption without sacrificing the security and accuracy of the model. The proposed solution addresses this challenge without modifying the protected DL model, thereby preserving the original prediction accuracy. Comprehensive experiments using different DL architectures and datasets demonstrate that inference services for large and complex DL models can be deployed in edge devices with TrustZone with limited trusted memory, ensuring data confidentiality and preserving the original model's prediction exactness.

Con2Mix (Contrastive Double Mixup) is a new semi-supervised learning methodology that innovates a triplet mixup data augmentation approach for finding code vulnerabilities in imbalanced, tabular security data sets. Tabular data sets in cybersecurity domains are widely known to pose challenges for machine learning because of their heavily imbalanced data (e.g., a small number of labeled attack samples buried in a sea of mostly benign, unlabeled data). Semi-supervised learning leverages a small subset of labeled data and a large subset of unlabeled data to train a learning model. While semi-supervised methods have been well studied in image and language domains, in security domains they remain underutilized, especially on tabular security data sets which pose especially difficult contextual information loss and balance challenges for machine learning. Experiments applying Con2Mix to collected security data sets show promise for addressing these challenges, achieving state-of-the-art performance on two evaluated data sets compared with other methods.

With the tremendous increase of the online data, training a single classifier may suffer because of the large variety of data domains. One solution could be to learn separate classifiers for each domain. However, this would arise a huge cost to gather annotated training data for a large number of domains and ignore similarity shared across domains. Hence, it leads to our problem setting: can labeled data from a related source domain help predict the unlabeled data in the target domain? In this paper, we consider two independent simultaneous data streams, which are referred to as the source and target streams. The target stream continuously generates data instances from one domain where the label is unknown, while the source stream continuously generates labeled data instances from another domain. Most likely, the two data streams would have different but related feature spaces and different data distributions. Moreover, these streams may have asynchronous concept drifts between them. Our problem setting, which is called Cross-domain Multistream Classification, is to predict the class labels of data instances in the target stream using a classifier trained on the labeled source stream. In this paper, we propose an efficient solution for cross-domain multistream classification by integrating change detection into online data stream adaptation. The class labels of data instances in the target stream are predicted using the sufficient amount of label information in the related source stream. And the concept drifts along the two independent streams are continuously being addressed at the same time. Experimental results on real-world data sets indicate significantly improved performance over baseline methods.