Fei WU Xinhao ZHENG Ying SUN Yang GAO Xiao-Yuan JING
Cross-project defect prediction (CPDP) is a hot research topic in recent years. The inconsistent data distribution between source and target projects and lack of labels for most of target instances bring a challenge for defect prediction. Researchers have developed several CPDP methods. However, the prediction performance still needs to be improved. In this paper, we propose a novel approach called Joint Domain Adaption and Pseudo-Labeling (JDAPL). The network architecture consists of a feature mapping sub-network to map source and target instances into a common subspace, followed by a classification sub-network and an auxiliary classification sub-network. The classification sub-network makes use of the label information of labeled instances to generate pseudo-labels. The auxiliary classification sub-network learns to reduce the distribution difference and improve the accuracy of pseudo-labels for unlabeled instances through loss maximization. Network training is guided by the adversarial scheme. Extensive experiments are conducted on 10 projects of the AEEEM and NASA datasets, and the results indicate that our approach achieves better performance compared with the baselines.
Jiao DU Shaojing FU Longjiang QU Chao LI Tianyin WANG Shanqi PANG
In this paper, by using the properties of the cyclic Hadamard matrices of order 4t, an infinite class of (4t-1)-variable 2-resilient rotation symmetric Boolean functions is constructed, and the nonlinearity of the constructed functions are also studied. To the best of our knowledge, this is the first class of direct constructions of 2-resilient rotation symmetric Boolean functions. The spirit of this method is different from the known methods depending on the solutions of an equation system proposed by Du Jiao, et al. Several situations are examined, as the direct corollaries, three classes of (4t-1)-variable 2-resilient rotation symmetric Boolean functions are proposed based on the corresponding sequences, such as m sequences, Legendre sequences, and twin primes sequences respectively.
Toshiro NAKAHIRA Koichi ISHIHARA Motoharu SASAKI Hirantha ABEYSEKERA Tomoki MURAKAMI Takatsune MORIYAMA Yasushi TAKATORI
In this paper, we propose a novel centralized control method to handle multi-radio and terminal connections in an 802.11ax wireless LAN (802.11ax) mixed environment. The proposed control method can improve the throughput by applying 802.11ax Spatial Reuse in an environment hosting different terminal standards and mixed terminal communication quality. We evaluate the proposed control method by computer simulations assuming environments with mixed terminal standards, mixed communication quality, and both.
Yasutaka MATSUDA Ryota SHIOYA Hideki ANDO
The high energy consumption of current processors causes several problems, including a limited clock frequency, short battery lifetime, and reduced device reliability. It is therefore important to reduce the energy consumption of the processor. Among resources in a processor, the issue queue (IQ) is a large consumer of energy, much of which is consumed by the wakeup logic. Within the wakeup logic, the tag comparison that checks source operand readiness consumes a significant amount of energy. This paper proposes an energy reduction scheme for tag comparison, called double-stage tag comparison. This scheme first compares the lower bits of the tag and then, only if these match, compares the higher bits. Because the energy consumption of tag comparison is roughly proportional to the total number of bits compared, energy is saved by reducing this number. However, this sequential comparison increases the delay of the IQ, thereby increasing the clock cycle time. Although this can be avoided by allocating an extra cycle to the issue operation, this in turn degrades the IPC. To avoid IPC degradation, we reconfigure a small number of entries in the IQ, where several oldest instructions that are likely to have an adverse effect on performance reside, to a single stage for tag comparison. Our evaluation results for SPEC2017 benchmark programs show that the double-stage tag comparison achieves on average a 21% reduction in the energy consumed by the wakeup logic (15% when including the overhead) with only 3.0% performance degradation.
Zhuotao LIAN Weizheng WANG Huakun HUANG Chunhua SU
In recent years, federated learning has attracted more and more attention as it could collaboratively train a global model without gathering the users' raw data. It has brought many challenges. In this paper, we proposed layer-based federated learning system with privacy preservation. We successfully reduced the communication cost by selecting several layers of the model to upload for global averaging and enhanced the privacy protection by applying local differential privacy. We evaluated our system in non independently and identically distributed scenario on three datasets. Compared with existing works, our solution achieved better performance in both model accuracy and training time.
Hiroki ISHIGURO Takashi ISHIDA Masashi SUGIYAMA
It has been demonstrated that large-scale labeled datasets facilitate the success of machine learning. However, collecting labeled data is often very costly and error-prone in practice. To cope with this problem, previous studies have considered the use of a complementary label, which specifies a class that an instance does not belong to and can be collected more easily than ordinary labels. However, complementary labels could also be error-prone and thus mitigating the influence of label noise is an important challenge to make complementary-label learning more useful in practice. In this paper, we derive conditions for the loss function such that the learning algorithm is not affected by noise in complementary labels. Experiments on benchmark datasets with noisy complementary labels demonstrate that the loss functions that satisfy our conditions significantly improve the classification performance.
Masayuki ODAGAWA Tetsushi KOIDE Toru TAMAKI Shigeto YOSHIDA Hiroshi MIENO Shinji TANAKA
This paper presents examination result of possibility for automatic unclear region detection in the CAD system for colorectal tumor with real time endoscopic video image. We confirmed that it is possible to realize the CAD system with navigation function of clear region which consists of unclear region detection by YOLO2 and classification by AlexNet and SVMs on customizable embedded DSP cores. Moreover, we confirmed the real time CAD system can be constructed by a low power ASIC using customizable embedded DSP cores.
Recently, control theory using machine learning, which is useful for the control of unknown systems, has attracted significant attention. This study focuses on such a topic with optimal control problems for unknown nonlinear systems. Because optimal controllers are designed based on mathematical models of the systems, it is challenging to obtain models with insufficient knowledge of the systems. Kernel functions are promising for developing data-driven models with limited knowledge. However, the complex forms of such kernel-based models make it difficult to design the optimal controllers. The design corresponds to solving Hamilton-Jacobi (HJ) equations because their solutions provide optimal controllers. Therefore, the aim of this study is to derive certain kernel-based models for which the HJ equations are solved in an exact sense, which is an extended version of the authors' former work. The HJ equations are decomposed into tractable algebraic matrix equations and nonlinear functions. Solving the matrix equations enables us to obtain the optimal controllers of the model. A numerical simulation demonstrates that kernel-based models and controllers are successfully developed.
Keitaro NAKASAI Masateru TSUNODA Kenichi MATSUMOTO
Software developers often use a web search engine to improve work efficiency. However, web search strategies (e.g., frequently changing web search keywords) may be different for each developer. In this study, we attempted to define a better web search strategy. Although many previous studies analyzed web search behavior in programming, they did not provide guidelines for web search strategies. To suggest guidelines for web search strategies, we asked 10 subjects four questions about programming which they had to solve, and analyzed their behavior. In the analysis, we focused on the subjects' task time and the web search metrics defined by us. Based on our experiment, to enhance the effectiveness of the search, we suggest (1) that one should not go through the next search result pages, (2) the number of keywords in queries should be suppressed, and (3) previously used keywords must be avoided when creating a new query.
We analyze the effect of window choice on the zero-padding method and corrected quadratically interpolated fast Fourier transform using a harmonic signal in noise at both high and low signal-to-noise ratios (SNRs) on a theoretical basis. Then, we validate the theoretical analysis using simulations. The theoretical analysis and simulation results using four traditional window functions show that the optimal window is determined depending on the SNR; the estimation errors are the smallest for the rectangular window at low SNR, the Hamming and Hanning windows at mid SNR, and the Blackman window at high SNR. In addition, we analyze the simulation results using the signal-to-noise floor ratio, which appears to be more effective than the conventional SNR in determining the optimal window.
Qin CHENG Linghua ZHANG Bo XUE Feng SHU Yang YU
As an emerging technology, device-free localization (DFL) using wireless sensor networks to detect targets not carrying any electronic devices, has spawned extensive applications, such as security safeguards and smart homes or hospitals. Previous studies formulate DFL as a classification problem, but there are still some challenges in terms of accuracy and robustness. In this paper, we exploit a generalized thresholding algorithm with parameter p as a penalty function to solve inverse problems with sparsity constraints for DFL. The function applies less bias to the large coefficients and penalizes small coefficients by reducing the value of p. By taking the distinctive capability of the p thresholding function to measure sparsity, the proposed approach can achieve accurate and robust localization performance in challenging environments. Extensive experiments show that the algorithm outperforms current alternatives.
Active network monitoring based on Boolean network tomography is a promising technique to localize link failures instantly in transport networks. However, the required set of monitoring trails must be recomputed after each link failure has occurred to handle succeeding link failures. Existing heuristic methods cannot compute the required monitoring trails in a sufficiently short time when multiple-link failures must be localized in the whole of large-scale managed networks. This paper proposes an approach for computing the required monitoring trails within an allowable expected period specified beforehand. A random walk-based analysis estimates the number of monitoring trails to be computed in the proposed approach. The estimated number of monitoring trails are computed by a lightweight method that only guarantees partial localization within restricted areas. The lightweight method is repeatedly executed until a successful set of monitoring trails achieving unambiguous localization in the entire managed networks can be obtained. This paper demonstrates that the proposed approach can compute a small number of monitoring trails for localizing all independent dual-link failures in managed networks made up of thousands of links within a given expected short period.
Osamu KAGAYA Keisuke ARAI Takato WATANABE Takuji ARIMA Toru UNO
In this paper, the influence of surface waves on the characteristics of on-glass antennas is clarified to enable appropriates design of C-band automotive on-glass antennas. Composite glasses are used in automotive windshields. These automotive composite glasses are composed of three layers. First, the surface wave properties of composite glass are investigated. Next, the effects of surface waves on the reflection coefficient characteristics of on-glass antennas are investigated. Finally, the antenna placement to reduce surface wave effect will be presented. Electromagnetic field analysis of a dipole antenna placed at the center of a 300mm × 300mm square flat composite glass showed that the electric field strength in the glass had ripples with the half wavelength period of the surface waves. Therefore, it was confirmed that standing waves are generated because of these surface waves. In addition, it is confirmed that ripples occur in the reflection coefficient at frequencies. Glass size is divisible by each of those guide wavelengths. Furthermore, it was clarified that the reflection coefficient fluctuates with respect to the distance between the antenna and a metal frame, which is attached to the end face in the direction perpendicular to the thickness of the glass because of the influence of standing waves caused by the surface waves; additionally, the reflection coefficient gets worse when the distance between the antenna and the metal frame is an integral multiple of one half wavelength. A similar tendency was observed in an electric field analysis using a model that was shaped like the actual windshield shape. Because radiation patterns also change as a result of the influence of surface waves and metal frames, the results imply that it is necessary to consider the actual device size and the metal frames when designing automotive on-glass antennas.
Van Hung PHAM Tuan Hung NGUYEN Duc Minh NGUYEN Hisashi MORISHITA
In this paper, we propose a new method based on copula theory to evaluate the detection performance of a distributed-processing multistatic radar system (DPMRS). By applying the Gaussian copula to model the dependence of local decisions in a DPMRS as well as data fusion rules of AND, OR, and K/N, the performance of a DPMRS for detecting Swerling fluctuating targets can be easily evaluated even under non-Gaussian clutter with a nonuniform dependence matrix. The reliability and flexibility of this method are validated by applying the proposed method to a previous problem by other authors, and our other investigation results indicate its high potential for evaluating DPMRS performance in various cases involving different models of target and clutter.
Aye Mon HTUN Maung SANN MAW Iwao SASASE P. Takis MATHIOPOULOS
In this paper, we propose a novel user selection scheme based on jointly combining channel gain (CG) and signal to interference plus noise ratio (SINR) to improve the sum-rate as well as to reduce the computation complexity of multi-user massive multi-input multi-output (MU-massive MIMO) downlink transmission through a block diagonalization (BD) precoding technique. By jointly considering CG and SINR based user sets, sum-rate performance improvement can be achieved by selecting higher gain users with better SINR conditions as well as by eliminating the users who cause low sum-rate in the system. Through this approach, the number of possible outcomes for the user selection scheme can be reduced by counting the common users for every pair of user combinations in the selection process since the common users of CG-based and SINR-based sets possess both higher channel gains and better SINR conditions. The common users set offers not only sum-rate performance improvements but also computation complexity reduction in the proposed scheme. It is shown by means of computer simulation experiments that the proposed scheme can increase the sum-rate with lower computation complexity for various numbers of users as compared to conventional schemes requiring the same or less computational complexity.
Yukasa MURAKAMI Masateru TSUNODA
Although many software engineering studies have been conducted, it is not clear whether they meet the needs of software development practitioners. Some studies evaluated the effectiveness of software engineering research by practitioners, to clarify the research satisfies the needs of the practitioners. We performed replicated study of them, recruiting practitioners who mainly belong to SMEs (small and medium-sized enterprises) to the survey. We asked 16 practitioners to evaluate cutting-edge software engineering studies presented in ICSE 2016. In the survey, we set the viewpoint of the evaluation as the effectiveness for the respondent's own work. As a result, the ratio of positive answers (i.e., the answers were greater than 2 on a 5-point scale) was 33.3%, and the ratio was lower than past studies. The result was not affected by the number of employees in the respondent's company, but would be affected by the viewpoint of the evaluation.
Kenshiro TAMATA Tomohiro MASHITA
A typical approach to reconstructing a 3D environment model is scanning the environment with a depth sensor and fitting the accumulated point cloud to 3D models. In this kind of scenario, a general 3D environment reconstruction application assumes temporally continuous scanning. However in some practical uses, this assumption is unacceptable. Thus, a point cloud matching method for stitching several non-continuous 3D scans is required. Point cloud matching often includes errors in the feature point detection because a point cloud is basically a sparse sampling of the real environment, and it may include quantization errors that cannot be ignored. Moreover, depth sensors tend to have errors due to the reflective properties of the observed surface. We therefore make the assumption that feature point pairs between two point clouds will include errors. In this work, we propose a feature description method robust to the feature point registration error described above. To achieve this goal, we designed a deep learning based feature description model that consists of a local feature description around the feature points and a global feature description of the entire point cloud. To obtain a feature description robust to feature point registration error, we input feature point pairs with errors and train the models with metric learning. Experimental results show that our feature description model can correctly estimate whether the feature point pair is close enough to be considered a match or not even when the feature point registration errors are large, and our model can estimate with higher accuracy in comparison to methods such as FPFH or 3DMatch. In addition, we conducted experiments for combinations of input point clouds, including local or global point clouds, both types of point cloud, and encoders.
Zihao SONG Peng SONG Chao SHENG Wenming ZHENG Wenjing ZHANG Shaokai LI
Unsupervised Feature selection is an important dimensionality reduction technique to cope with high-dimensional data. It does not require prior label information, and has recently attracted much attention. However, it cannot fully utilize the discriminative information of samples, which may affect the feature selection performance. To tackle this problem, in this letter, we propose a novel discriminative virtual label regression method (DVLR) for unsupervised feature selection. In DVLR, we develop a virtual label regression function to guide the subspace learning based feature selection, which can select more discriminative features. Moreover, a linear discriminant analysis (LDA) term is used to make the model be more discriminative. To further make the model be more robust and select more representative features, we impose the ℓ2,1-norm on the regression and feature selection terms. Finally, extensive experiments are carried out on several public datasets, and the results demonstrate that our proposed DVLR achieves better performance than several state-of-the-art unsupervised feature selection methods.
Bodin CHINTHANET Raula GAIKOVINA KULA Rodrigo ELIZA ZAPATA Takashi ISHIO Kenichi MATSUMOTO Akinori IHARA
It has become common practice for software projects to adopt third-party dependencies. Developers are encouraged to update any outdated dependency to remain safe from potential threats of vulnerabilities. In this study, we present an approach to aid developers show whether or not a vulnerable code is reachable for JavaScript projects. Our prototype, SōjiTantei, is evaluated in two ways (i) the accuracy when compared to a manual approach and (ii) a larger-scale analysis of 780 clients from 78 security vulnerability cases. The first evaluation shows that SōjiTantei has a high accuracy of 83.3%, with a speed of less than a second analysis per client. The second evaluation reveals that 68 out of the studied 78 vulnerabilities reported having at least one clean client. The study proves that automation is promising with the potential for further improvement.
Huy H. NGUYEN Minoru KURIBAYASHI Junichi YAMAGISHI Isao ECHIZEN
Deep neural networks (DNNs) have achieved excellent performance on several tasks and have been widely applied in both academia and industry. However, DNNs are vulnerable to adversarial machine learning attacks in which noise is added to the input to change the networks' output. Consequently, DNN-based mission-critical applications such as those used in self-driving vehicles have reduced reliability and could cause severe accidents and damage. Moreover, adversarial examples could be used to poison DNN training data, resulting in corruptions of trained models. Besides the need for detecting adversarial examples, correcting them is important for restoring data and system functionality to normal. We have developed methods for detecting and correcting adversarial images that use multiple image processing operations with multiple parameter values. For detection, we devised a statistical-based method that outperforms the feature squeezing method. For correction, we devised a method that uses for the first time two levels of correction. The first level is label correction, with the focus on restoring the adversarial images' original predicted labels (for use in the current task). The second level is image correction, with the focus on both the correctness and quality of the corrected images (for use in the current and other tasks). Our experiments demonstrated that the correction method could correct nearly 90% of the adversarial images created by classical adversarial attacks and affected only about 2% of the normal images.