Biblio

Recent breakthrough in neural network has led to the birth of Convolutional neural network (CNN) which has been found to be very efficient especially in the areas of image recognition and classification. This success is traceable to the availability of large datasets and its capability to learn salient and complex data features which subsequently produce a reusable output model (Fθ). The Fθ are often made available (e.g. on cloud as-a-service) for others (client) to train their data or do transfer learning, however, an adversary can perpetrate a model inversion attack on the model Fθ to recover training data, hence compromising the sensitivity of the model buildup data. This is possible because CNN as a variant of deep neural network does memorize most of its training data during learning. Consequently, this has pose a privacy concern especially when a medical or financial data are used as model buildup data. Existing researches that proffers privacy preserving approach however suffer from significant accuracy degradation and this has left privacy preserving model on a theoretical desk. In this paper, we proposed an ϵ-tuple differential privacy approach that is based on neuron impact factor estimation to preserve privacy of CNN model without significant accuracy degradation. We experiment our approach on two large datasets and the result shows no significant accuracy degradation.

The anomaly detection of time series is a hotspot of time series data mining. The own characteristics of different anomaly detectors determine the abnormal data that they are good at. There is no detector can be optimizing in all types of anomalies. Moreover, it still has difficulties in industrial production due to problems such as a single detector can't be optimized at different time windows of the same time series. This paper proposes an adaptive model based on time series characteristics and selecting appropriate detector and run-time parameters for anomaly detection, which is called ATSDLN(Adaptive Time Series Detector Learning Network). We take the time series as the input of the model, and learn the time series representation through FCN. In order to realize the adaptive selection of detectors and run-time parameters according to the input time series, the outputs of FCN are the inputs of two sub-networks: the detector selection network and the run-time parameters selection network. In addition, the way that the variable layer width design of the parameter selection sub-network and the introduction of transfer learning make the model be with more expandability. Through experiments, it is found that ATSDLN can select appropriate anomaly detector and run-time parameters, and have strong expandability, which can quickly transfer. We investigate the performance of ATSDLN in public data sets, our methods outperform other methods in most cases with higher effect and better adaptation. We also show experimental results on public data sets to demonstrate how model structure and transfer learning affect the effectiveness.

Corpora used to learn open-domain Question-Answering (QA) models are typically collected from a wide variety of topics or domains. Since QA requires understanding natural language, open-domain QA models generally need very large training corpora. A simple way to alleviate data demand is to restrict the domain covered by the QA model, leading thus to domain-specific QA models. While learning improved QA models for a specific domain is still challenging due to the lack of sufficient training data in the topic of interest, additional training data can be obtained from related topic domains. Thus, instead of learning a single open-domain QA model, we investigate domain adaptation approaches in order to create multiple improved domain-specific QA models. We demonstrate that this can be achieved by stratifying the source dataset, without the need of searching for complementary data unlike many other domain adaptation approaches. We propose a deep architecture that jointly exploits convolutional and recurrent networks for learning domain-specific features while transferring domain-shared features. That is, we use transferable features to enable model adaptation from multiple source domains. We consider different transference approaches designed to learn span-level and sentence-level QA models. We found that domain-adaptation greatly improves sentence-level QA performance, and span-level QA benefits from sentence information. Finally, we also show that a simple clustering algorithm may be employed when the topic domains are unknown and the resulting loss in accuracy is negligible.

In recent years, deep convolution neural networks (DCNNs) have won many contests in machine learning, object detection, and pattern recognition. Furthermore, deep learning techniques achieved exceptional performance in image classification, reaching accuracy levels beyond human capability. Malware variants from similar categories often contain similarities due to code reuse. Converting malware samples into images can cause these patterns to manifest as image features, which can be exploited for DCNN classification. Techniques for converting malware binaries into images for visualization and classification have been reported in the literature, and while these methods do reach a high level of classification accuracy on training datasets, they tend to be vulnerable to overfitting and perform poorly on previously unseen samples. In this paper, we explore and document a variety of techniques for representing malware binaries as images with the goal of discovering a format best suited for deep learning. We implement a database for malware binaries from several families, stored in hexadecimal format. These malware samples are converted into images using various approaches and are used to train a neural network to recognize visual patterns in the input and classify malware based on the feature vectors. Each image type is assessed using a variety of learning models, such as transfer learning with existing DCNN architectures and feature extraction for support vector machine classifier training. Each technique is evaluated in terms of classification accuracy, result consistency, and time per trial. Our preliminary results indicate that improved image representation has the potential to enable more effective classification of new malware.

Despite several attacks have been proposed, text-based CAPTCHAs are still being widely used as a security mechanism. One of the reasons for the pervasive use of text captchas is that many of the prior attacks are scheme-specific and require a labor-intensive and time-consuming process to construct. This means that a change in the captcha security features like a noisier background can simply invalid an earlier attack. This paper presents a generic, yet effective text captcha solver based on the generative adversarial network. Unlike prior machine-learning-based approaches that need a large volume of manually-labeled real captchas to learn an effective solver, our approach requires significantly fewer real captchas but yields much better performance. This is achieved by first learning a captcha synthesizer to automatically generate synthetic captchas to learn a base solver, and then fine-tuning the base solver on a small set of real captchas using transfer learning. We evaluate our approach by applying it to 33 captcha schemes, including 11 schemes that are currently being used by 32 of the top-50 popular websites including Microsoft, Wikipedia, eBay and Google. Our approach is the most capable attack on text captchas seen to date. It outperforms four state-of-the-art text-captcha solvers by not only delivering a significant higher accuracy on all testing schemes, but also successfully attacking schemes where others have zero chance. We show that our approach is highly efficient as it can solve a captcha within 0.05 second using a desktop GPU. We demonstrate that our attack is generally applicable because it can bypass the advanced security features employed by most modern text captcha schemes. We hope the results of our work can encourage the community to revisit the design and practical use of text captchas.

Here we explore the applicability of traditional sliding window based convolutional neural network (CNN) detection pipeline and region based object detection techniques such as Faster Region-based CNN (R-CNN) and Region-based Fully Convolutional Networks (R-FCN) on the problem of object detection in X-ray security imagery. Within this context, with limited dataset availability, we employ a transfer learning paradigm for network training tackling both single and multiple object detection problems over a number of R-CNN/R-FCN variants. The use of first-stage region proposal within the Faster RCNN and R-FCN provide superior results than traditional sliding window driven CNN (SWCNN) approach. With the use of Faster RCNN with VGG16, pretrained on the ImageNet dataset, we achieve 88.3 mAP for a six object class X-ray detection problem. The use of R-FCN with ResNet-101, yields 96.3 mAP for the two class firearm detection problem requiring 0.1 second computation per image. Overall we illustrate the comparative performance of these techniques as object localization strategies within cluttered X-ray security imagery.

There have been a growing number of interests in using the convolutional neural network(CNN) in image forensics, where some excellent methods have been proposed. Training the randomly initialized model from scratch needs a big amount of training data and computational time. To solve this issue, we present a new method of training an image forensic model using prior knowledge transferred from the existing steganalysis model. We also find out that CNN models tend to show poor performance when tested on a different database. With knowledge transfer, we are able to easily train an excellent model for a new database with a small amount of training data from the new database. Performance of our models are evaluated on Bossbase and BOW by detecting five forensic types, including median filtering, resampling, JPEG compression, contrast enhancement and additive Gaussian noise. Through a series of experiments, we demonstrate that our proposed method is very effective in two scenario mentioned above, and our method based on transfer learning can greatly accelerate the convergence of CNN model. The results of these experiments show that our proposed method can detect five different manipulations with an average accuracy of 97.36%.

Deep learning techniques have demonstrated the ability to perform a variety of object recognition tasks using visible imager data; however, deep learning has not been implemented as a means to autonomously detect and assess targets of interest in a physical security system. We demonstrate the use of transfer learning on a convolutional neural network (CNN) to significantly reduce training time while keeping detection accuracy of physical security relevant targets high. Unlike many detection algorithms employed by video analytics within physical security systems, this method does not rely on temporal data to construct a background scene; targets of interest can halt motion indefinitely and still be detected by the implemented CNN. A key advantage of using deep learning is the ability for a network to improve over time. Periodic retraining can lead to better detection and higher confidence rates. We investigate training data size versus CNN test accuracy using physical security video data. Due to the large number of visible imagers, significant volume of data collected daily, and currently deployed human in the loop ground truth data, physical security systems present a unique environment that is well suited for analysis via CNNs. This could lead to the creation of algorithmic element that reduces human burden and decreases human analyzed nuisance alarms.

New and unseen network attacks pose a great threat to the signature-based detection systems. Consequently, machine learning-based approaches are designed to detect attacks, which rely on features extracted from network data. The problem is caused by different distribution of features in the training and testing datasets, which affects the performance of the learned models. Moreover, generating labeled datasets is very time-consuming and expensive, which undercuts the effectiveness of supervised learning approaches. In this paper, we propose using transfer learning to detect previously unseen attacks. The main idea is to learn the optimized representation to be invariant to the changes of attack behaviors from labeled training sets and non-labeled testing sets, which contain different types of attacks and feed the representation to a supervised classifier. To the best of our knowledge, this is the first effort to use a feature-based transfer learning technique to detect unseen variants of network attacks. Furthermore, this technique can be used with any common base classifier. We evaluated the technique on publicly available datasets, and the results demonstrate the effectiveness of transfer learning to detect new network attacks.

Several defect prediction models proposed are effective when historical datasets are available. Defect prediction becomes difficult when no historical data exist. Cross-project defect prediction (CPDP), which uses projects from other sources/companies to predict the defects in the target projects proposed in recent studies has shown promising results. However, the performance of most CPDP approaches are still beyond satisfactory mainly due to distribution mismatch between the source and target projects. In this study, a credibility theory based Naïve Bayes (CNB) classifier is proposed to establish a novel reweighting mechanism between the source projects and target projects so that the source data could simultaneously adapt to the target data distribution and retain its own pattern. Our experimental results show that the feasibility of the novel algorithm design and demonstrate the significant improvement in terms of the performance metrics considered achieved by CNB over other CPDP approaches.