Due to the superior performance, deep learning has been widely applied to various applications, including image classification, bioinformatics, and cybersecurity. Nevertheless, the research investigations on deep learning in the adversarial environment are still on their preliminary stage. The emerging adversarial learning methods, e.g., generative adversarial networks, have introduced two vital questions: to what degree the security of deep learning with the presence of adversarial examples is; how to evaluate the performance of deep learning models in adversarial environment, thus, to raise security advice such that the selected application system based on deep learning is resistant to adversarial examples. To see the answers, we leverage image classification as an example application scenario to propose a framework of Evaluating Deep Learning for Image Classification (EDLIC) to conduct comprehensively quantitative analysis. Moreover, we introduce a set of evaluating metrics to measure the performance of different attacking and defensive techniques. After that, we conduct extensive experiments towards the performance of deep learning for image classification under different adversarial environments to validate the scalability of EDLIC. Finally, we give some advice about the selection of deep learning models for image classification based on these comparative results.

This paper analyzes the optimal SSD configuration including emerging non-volatile memories such as quadruple-level cell (QLC) NAND flash memory [1] and storage class memories (SCMs). First, SSD performance and SSD endurance lifetime of hybrid SSD are evaluated in four configurations: 1) single-level cell (SLC)/QLC NAND flash, 2) SCM/QLC NAND flash, 3) SCM/triple-level cell (TLC)/QLC NAND flash and 4) SCM/TLC NAND flash. Furthermore, these four configurations are compared in limited cost. In case of cold workloads or high total SSD cost assumption, SCM/TLC NAND flash hybrid configuration is recommended in both SSD performance and endurance lifetime. For hot workloads with low total SSD cost assumption, however, SLC/QLC NAND flash hybrid configuration is recommended with emphasis on SSD endurance lifetime. Under the same conditions as above, SCM/TLC/QLC NAND flash tri-hybrid is the best configuration in SSD performance considering cost. In particular, for prxy_0 (write-hot workload), SCM/TLC/QLC NAND flash tri-hybrid achieves 67% higher IOPS/cost than SCM/TLC NAND flash hybrid. Moreover, the configurations with the highest IOPS/cost in each workload and cost limit are picked up and analyzed with various types of SCMs. For all cases except for the case of prxy_1 with high total SSD cost assumption, middle-end SCM (write latency: 1us, read latency: 1us) is recommended in performance considering cost. However, for prxy_1 (read-hot workload) with high total SSD cost assumption, high-end SCM (write latency: 100ns, read latency: 100ns) achieves the best performance.

The outcome of document clustering depends on the scheme used to assign a weight to each term in a document. While recent works have tried to use distributions related to class to enhance the discrimination ability. It is worth exploring whether a deviation approach or an entropy approach is more effective. This paper presents a comparison between deviation-based distribution and entropy-based distribution as constraints in term weighting. In addition, their potential combinations are investigated to find optimal solutions in guiding the clustering process. In the experiments, the seeded k-means method is used for clustering, and the performances of deviation-based, entropy-based, and hybrid approaches, are analyzed using two English and one Thai text datasets. The result showed that the deviation-based distribution outperformed the entropy-based distribution, and a suitable combination of these distributions increases the clustering accuracy by 10%.

This paper analyzes the system-level performance of Storage Class Memory (SCM)/NAND flash hybrid solid-state drives (SSDs) and SCM/NAND flash/NAND flash tri-hybrid SSDs in difference types of NAND flash memory. There are several types of NAND flash memory, i.e. 2-dimensional (2D) or 3-dimensional (3D), charge-trap type (CT) and floating-gate type (FG) and multi-level cell (MLC) or triple-level cell (TLC). In this paper, the following four types of NAND flash memory are analyzed: 1) 3D CT TLC, 2) 3D FG TLC, 3) 2D FG TLC, and 4) 2D FG MLC NAND flash. Regardless of read- and write-intensive workloads, SCM/NAND flash hybrid SSD with low cost 3D CT TLC NAND flash achieves the best performance that is 20% higher than that with higher cost 2D FG MLC NAND flash. The performance improvement of 3D CT TLC NAND flash can be obtained by the short write latency. On the other hand, in case of tri-hybrid SSD, SCM/3D CT TLC/3D CT TLC NAND flash tri-hybrid SSD improves the performance 102% compared to SCM/2D FG MLC/3D CT TLC NAND flash tri-hybrid SSD. In addition, SCM/2D FG MLC/2D FG MLC NAND flash tri-hybrid SSD shows 49% lower performance than SCM/2D FG MLC/3D CT TLC NAND flash tri-hybrid SSD. Tri-hybrid SSD flash with 3D CT TLC NAND flash is the best performance in tri-hybrid SSD thanks to larger block size and word-line (WL) write. Therefore, in 3D CT TLC NAND flash based SSDs, higher cost MLC NAND flash is not necessary for hybrid SSD and tri-hybrid SSD for data center applications.

Recent years have witnessed a rapid increase in cyber-attacks through unauthorized accesses and DDoS attacks. Since packet classification is a fundamental technique to prevent such illegal communications, it has gained considerable attention. Packet classification is achieved with a linear search on a classification rule list that represents the packet classification policy. As such, a large number of rules can result in serious communication latency. To decrease this latency, the problem is formalized as optimal rule ordering (ORO). In most cases, this problem aims to find the order of rules that minimizes latency while satisfying the dependency relation of the rules, where rules ri and rj are dependent if there is a packet that matches both ri and rj and their actions applied to packets are different. However, there is a case in which although the ordering violates the dependency relation, the ordering satisfies the packet classification policy. Since such an ordering can decrease the latency compared to an ordering under the constraint of the dependency relation, we have introduced a new model, called relaxed optimal rule ordering (RORO). In general, it is difficult to determine whether an ordering satisfies the classification policy, even when it violates the dependency relation, because this problem contains unsatisfiability. However, using a zero-suppressed binary decision diagram (ZDD), we can determine it in a reasonable amount of time. In this paper, we present a simulated annealing method for RORO which interchanges rules by determining whether rules ri and rj can be interchanged in terms of policy violation using the ZDD. The experimental results show that our method decreases latency more than other heuristics.

Music classification has been inspired by the remarkable success of deep learning. To enhance efficiency and ensure high performance at the same time, a hybrid architecture that combines deep learning and Broad Learning (BL) is proposed for music classification tasks. At the feature extraction stage, the Random CNN (RCNN) is adopted to analyze the Mel-spectrogram of the input music sound. Compared with conventional CNN, RCNN has more flexible structure to adapt to the variance contained in different types of music. At the prediction stage, the BL technique is introduced to enhance the prediction accuracy and reduce the training time as well. Experimental results on three benchmark datasets (GTZAN, Ballroom, and Emotion) demonstrate that: i) The proposed scheme achieves higher classification accuracy than the deep learning based one, which combines CNN and LSTM, on all three benchmark datasets. ii) Both RCNN and BL contribute to the performance improvement of the proposed scheme. iii) The introduction of BL also helps to enhance the prediction efficiency of the proposed scheme.

Inspired by the efficient proof procedures discussed in Computability logic [3],[5],[6], we describe a heuristic proof procedure for first-order logic. This is a variant of Gentzen sequent system [2] and has the following features: (a) it views sequents as games between the machine and the environment, and (b) it views proofs as a winning strategy of the machine. From this game-based viewpoint, a poweful heuristic can be extracted and a fair degree of determinism in proof search can be obtained. This article proposes a new deductive system LKg with respect to first-order logic and proves its soundness and completeness.

Threat object recognition in x-ray security images is one of the important practical applications of computer vision. However, research in this field has been limited by the lack of available dataset that would mirror the practical setting for such applications. In this paper, we present a novel GAN-based anomaly detection (GBAD) approach as a solution to the extreme class-imbalance problem in multi-label classification. This method helps in suppressing the surge in false positives induced by training a CNN on a non-practical dataset. We evaluate our method on a large-scale x-ray image database to closely emulate practical scenarios in port security inspection systems. Experiments demonstrate improvement against the existing algorithm.

In this work, we present one novel rust detection method based upon one-class classification and L2 sparse representation (SR) with decision fusion. Firstly, a new color contrast descriptor is proposed for extracting the rust features of steel structure images. Considering that the patterns of rust features are more simplified than those of non-rust ones, one-class support vector machine (SVM) classifier and L2 SR classifier are designed with these rust image features, respectively. After that, a multiplicative fusion rule is advocated for combining the one-class SVM and L2 SR modules, thereby achieving more accurate rust detecting results. In the experiments, we conduct numerous experiments, and when compared with other developed rust detectors, the presented method can offer better rust detecting performances.

This work develops a system called CLAP that detects and classifies “potentially unwanted applications” (PUAs) such as adware or remote monitoring tools. Our approach leverages DNS queries made by apps. Using a large sample of Android apps from third-party marketplaces, we first reveal that DNS queries can provide useful information for detection and classification of PUAs. We then show that existing DNS blacklists are limited when performing these tasks. Finally, we demonstrate that the CLAP system performs with high accuracy.

A drastic increase in cyberattacks targeting Internet of Things (IoT) devices using telnet protocols has been observed. IoT malware continues to evolve, and the diversity of OS and environments increases the difficulty of executing malware samples in an observation setting. To address this problem, we sought to develop an alternative means of investigation by using the telnet logs of IoT honeypots and analyzing malware without executing it. In this paper, we present a malware classification method based on malware binaries, command sequences, and meta-features. We employ both unsupervised or supervised learning algorithms and text-mining algorithms for handling unstructured data. Clustering analysis is applied for finding malware family members and revealing their inherent features for better explanation. First, the malware binaries are grouped using similarity analysis. Then, we extract key patterns of interaction behavior using an N-gram model. We also train a multiclass classifier to identify IoT malware categories based on common infection behavior. For misclassified subclasses, second-stage sub-training is performed using a file meta-feature. Our results demonstrate 96.70% accuracy, with high precision and recall. The clustering results reveal variant attack vectors and one denial of service (DoS) attack that used pure Linux commands.

In this paper, we study domain adaptation of semantic role classification. Most systems utilize the supervised method for semantic role classification. But, these methods often suffer severe performance drops on out-of-domain test data. The reason for the performance drops is that there are giant feature differences between source and target domain. This paper proposes a framework called Adversarial Domain Adaption Network (ADAN) to relieve domain adaption of semantic role classification. The idea behind our method is that the proposed framework can derive domain-invariant features via adversarial learning and narrow down the gap between source and target feature space. To evaluate our method, we conduct experiments on English portion in the CoNLL 2009 shared task. Experimental results show that our method can largely reduce the performance drop on out-of-domain test data.

Clinical researches show that the morbidity of coronary artery disease (CAD) is gradually increasing in many countries every year, and it causes hundreds of thousands of people all over the world dying for each year. As the optical coherence tomography with high resolution and better contrast applied to the lesion tissue investigation of human vessel, many more micro-structures of the vessel could be easily and clearly visible to doctors, which help to improve the CAD treatment effect. Manual qualitative analysis and classification of vessel lesion tissue are time-consuming to doctors because a single-time intravascular optical coherence (IVOCT) data set of a patient usually contains hundreds of in-vivo vessel images. To overcome this problem, we focus on the investigation of the superficial layer of the lesion region and propose a model based on local multi-layer region for vessel lesion components (lipid, fibrous and calcified plaque) features characterization and extraction. At the pre-processing stage, we applied two novel automatic methods to remove the catheter and guide-wire respectively. Based on the detected lumen boundary, the multi-layer model in the proximity lumen boundary region (PLBR) was built. In the multi-layer model, features extracted from the A-line sub-region (ALSR) of each layer was employed to characterize the type of the tissue existing in the ALSR. We used 7 human datasets containing total 490 OCT images to assess our tissue classification method. Validation was obtained by comparing the manual assessment with the automatic results derived by our method. The proposed automatic tissue classification method achieved an average accuracy of 89.53%, 93.81% and 91.78% for fibrous, calcified and lipid plaque respectively.

Since the development of food diaries could enable people to develop healthy eating habits, food image recognition is in high demand to reduce the effort in food recording. Previous studies have worked on this challenging domain with datasets having fixed numbers of samples and classes. However, in the real-world setting, it is impossible to include all of the foods in the database because the number of classes of foods is large and increases continually. In addition to that, inter-class similarity and intra-class diversity also bring difficulties to the recognition. In this paper, we solve these problems by using deep convolutional neural network features to build a personalized classifier which incrementally learns the user's data and adapts to the user's eating habit. As a result, we achieved the state-of-the-art accuracy of food image recognition by the personalization of 300 food records per user.

In recent decade, many state-of-the-art algorithms on image classification as well as audio classification have achieved noticeable successes with the development of deep convolutional neural network (CNN). However, most of the works only exploit single type of training data. In this paper, we present a study on classifying bird species by exploiting the combination of both visual (images) and audio (sounds) data using CNN, which has been sparsely treated so far. Specifically, we propose CNN-based multimodal learning models in three types of fusion strategies (early, middle, late) to settle the issues of combining training data cross domains. The advantage of our proposed method lies on the fact that we can utilize CNN not only to extract features from image and audio data (spectrogram) but also to combine the features across modalities. In the experiment, we train and evaluate the network structure on a comprehensive CUB-200-2011 standard data set combing our originally collected audio data set with respect to the data species. We observe that a model which utilizes the combination of both data outperforms models trained with only an either type of data. We also show that transfer learning can significantly increase the classification performance.

Along with the proliferation of IoT (Internet of Things) devices, cyberattacks towards them are on the rise. In this paper, aiming at efficient precaution and mitigation of emerging IoT cyberthreats, we present a multimodal study on applying machine learning methods to characterize malicious programs which target multiple IoT platforms. Experiments show that opcode sequences obtained from static analysis and API sequences obtained by dynamic analysis provide sufficient discriminant information such that IoT malware can be classified with near optimal accuracy. Automated and accelerated identification and mitigation of new IoT cyberthreats can be enabled based on the findings reported in this study.

The packet classification problem to determine the behavior of incoming packets at the network devices. The processing latency of packet classification by linear search is proportional to the number of classification rules. To limit the latency caused by classification to a certain level, we should develop a classification algorithm that classifies packets in a time independent of the number of classification rules. Arbitrary (including noncontiguous) bitmask rules are efficiently expressive for controlling higher layer communication, achiving access control lists, Quality of Service and so on. In this paper, we propose a classification algorithm based on run-based trie [1] according to arbitrary bitmask rules. The space complexity of proposed algorithm is in linear in the size of a rule list. The time complexity except for construction of that can be regarded as constant which is independent the number of rules. Experimental results using a packet classification algorithm benchmark [2] show that our method classifies packets in constant time independent of the number of rules.

We propose a class-incremental learning framework for human activity recognition based on the Bag-of-Sequencelets model (BoS). The framework updates learned models efficiently without having to relearn them when training data of new classes are added. In this framework, all types of features including hand-crafted features and Convolutional Neural Networks (CNNs) based features and combinations of those features can be used as features for videos. Compared with the original BoS, the new framework can reduce the learning time greatly with little loss of classification accuracy.

It is not easy for a student to present a question or comment to the lecturer and other students in large classes. This paper introduces a new audience presentation system (APS), which creates slide presentations of students' mobile responses in the classroom. Experimental surveys demonstrate the utility of this APS for classroom interactivity.

In this research, we propose a novel method to determine fingerprint liveness to improve the discriminative behavior and classification accuracy of the combined features. This approach detects if a fingerprint is from a live or fake source. In this approach, fingerprint images are analyzed in the differential excitation (DE) component and the centralized binary pattern (CBP) component, which yield the DE image and CBP image, respectively. The images obtained are used to generate a two-dimensional histogram that is subsequently used as a feature vector. To decide if a fingerprint image is from a live or fake source, the feature vector is processed using support vector machine (SVM) classifiers. To evaluate the performance of the proposed method and compare it to existing approaches, we conducted experiments using the datasets from the 2011 and 2015 Liveness Detection Competition (LivDet), collected from four sensors. The results show that the proposed method gave comparable or even better results and further prove that methods derived from combination of features provide a better performance than existing methods.