Jaeseong JEONG Chang Heon KIM Namhun KOO Soonhak KWON Sumin LEE
The differential uniformity, the boomerang uniformity, and the extended Walsh spectrum etc are important parameters to evaluate the security of S (substitution)-box. In this paper, we introduce efficient formulas to compute these cryptographic parameters of permutation polynomials of the form xrh(x(2n-1)/d) over a finite field of q=2n elements, where r is a positive integer and d is a positive divisor of 2n-1. The computational cost of those formulas is proportional to d. We investigate differentially 4-uniform permutation polynomials of the form xrh(x(2n-1)/3) and compute the boomerang spectrum and the extended Walsh spectrum of them using the suggested formulas when 6≤n≤12 is even, where d=3 is the smallest nontrivial d for even n. We also investigate the differential uniformity of some permutation polynomials introduced in some recent papers for the case d=2n/2+1.
Lu ZHANG Chengqun WANG Mengyuan FANG Weiqiang XU
To solve the problem of metamerism in the color reproduction process, various spectral reflectance reconstruction methods combined with neural network have been proposed in recent years. However, these methods are generally sensitive to initial values and can easily converge to local optimal solutions, especially on small data sets. In this paper, we propose a spectral reflectance reconstruction algorithm based on the Back Propagation Neural Network (BPNN) and an improved Sparrow Search Algorithm (SSA). In this algorithm, to solve the problem that BPNN is sensitive to initial values, we propose to use SSA to initialize BPNN, and we use the sine chaotic mapping to further improve the stability of the algorithm. In the experiment, we tested the proposed algorithm on the X-Rite ColorChecker Classic Mini Chart which contains 24 colors, the results show that the proposed algorithm has significantly better performance compared to other algorithms and moreover it can meet the needs of spectral reflectance reconstruction on small data sets. Code is avaible at https://github.com/LuraZhang/spectral-reflectance-reconsctuction.
Tao YU Yang YANG Hua MENG Yong WANG
Almost-complementary pairs (ACPs) are sequence pairs whose autocorrelations sum up to zero at all but one non-zero time-shifts. Periodic ACPs (P-ACPs) display almost similar correlation properties to that of the periodic complementary pairs (PCPs). In this letter, we propose systematic constructions of quadriphase P-ACPs (QP-ACPs) from aperiodic (periodic) complementary pairs and almost perfect binary (quadriphase) sequences. The proposed construction gives QP-ACPs of new lengths which are not covered in the literature.
The application of compressed sensing (CS) theory to non-orthogonal multiple access (NOMA) systems has been investigated recently. As described in this paper, we propose a quality-of-service (QoS)-aware, low-complexity, CS-based user selection and power allocation scheme with adaptive resource block selection for downlink NOMA systems, where the tolerable interference threshold is designed mathematically to achieve a given QoS requirement by being relaxed to a constrained l1 norm optimization problem. The proposed scheme adopts two adaptive resource block (RB) selection algorithms that assign proper RB to user pairs, i.e. max-min channel assignment and two-step opportunistic channel assignment. Simulation results show that the proposed scheme is more effective at improving the user rate than other reference schemes while reducing the required complexity. The QoS requirement is approximately satisfied as long as the required QoS value is feasible.
Noninvasive recognition is an important trend in diabetes recognition. Unfortunately, the accuracy obtained from the conventional noninvasive recognition methods is low. This paper proposes a novel Diabetes Noninvasive Recognition method via the plantar pressure image and improved Capsule Network (DNR-CapsNet). The input of the proposed method is a plantar pressure image, and the output is the recognition result: healthy or possibly diabetes. The ResNet18 is used as the backbone of the convolutional layers to convert pixel intensities to local features in the proposed DNR-CapsNet. Then, the PrimaryCaps layer, SecondaryCaps layer, and DiabetesCaps layer are developed to achieve the diabetes recognition. The semantic fusion and locality-constrained dynamic routing are also developed to further improve the recognition accuracy in our method. The experimental results indicate that the proposed method has a better performance on diabetes noninvasive recognition than the state-of-the-art methods.
Anna HIRAI Yuichi MATSUMOTO Takanori SATO Tadashi KAWAI Akira ENOKIHARA Shinya NAKAJIMA Atsushi KANNO Naokatsu YAMAMOTO
A Mach-Zehnder optical modulator with the tunable multimode interference coupler was fabricated using Ti-diffused LiNbO3. The modulation extinction ratio could be voltage controlled to maximize up to 50 dB by tuning the coupler. Optical single-sideband modulation was also achieved with a sideband suppression ratio of more than 30 dB.
Yi LIU Wei QIN Qibin ZHENG Gensong LI Mengmeng LI
Feature selection based on particle swarm optimization is often employed for promoting the performance of artificial intelligence algorithms. However, its interpretability has been lacking of concrete research. Improving the stability of the feature selection method is a way to effectively improve its interpretability. A novel feature selection approach named Interpretable Particle Swarm Optimization is developed in this paper. It uses four data perturbation ways and three filter feature selection methods to obtain stable feature subsets, and adopts Fuch map to convert them to initial particles. Besides, it employs similarity mutation strategy, which applies Tanimoto distance to choose the nearest 1/3 individuals to the previous particles to implement mutation. Eleven representative algorithms and four typical datasets are taken to make a comprehensive comparison with our proposed approach. Accuracy, F1, precision and recall rate indicators are used as classification measures, and extension of Kuncheva indicator is employed as the stability measure. Experiments show that our method has a better interpretability than the compared evolutionary algorithms. Furthermore, the results of classification measures demonstrate that the proposed approach has an excellent comprehensive classification performance.
Hongzhe LIU Ningwei WANG Xuewei LI Cheng XU Yaze LI
In the neck part of a two-stage object detection network, feature fusion is generally carried out in either a top-down or bottom-up manner. However, two types of imbalance may exist: feature imbalance in the neck of the model and gradient imbalance in the region of interest extraction layer due to the scale changes of objects. The deeper the network is, the more abstract the learned features are, that is to say, more semantic information can be extracted. However, the extracted image background, spatial location, and other resolution information are less. In contrast, the shallow part can learn little semantic information, but a lot of spatial location information. We propose the Both Ends to Centre to Multiple Layers (BEtM) feature fusion method to solve the feature imbalance problem in the neck and a Multi-level Region of Interest Feature Extraction (MRoIE) layer to solve the gradient imbalance problem. In combination with the Region-based Convolutional Neural Network (R-CNN) framework, our Balanced Feature Fusion (BFF) method offers significantly improved network performance compared with the Faster R-CNN architecture. On the MS COCO 2017 dataset, it achieves an average precision (AP) that is 1.9 points and 3.2 points higher than those of the Feature Pyramid Network (FPN) Faster R-CNN framework and the Generic Region of Interest Extractor (GRoIE) framework, respectively.
Xianmei FANG Xiaobo GAO Yuting WANG Zhouyu LIAO Yue MA
Fault localization analyzes the runtime information of two classes of test cases (i.e., passing test cases and failing test cases) to identify suspicious statements potentially responsible for a failure. However, the failing test cases are always far fewer than passing test cases in reality, and the class imbalance problem will affect fault localization effectiveness. To address this issue, we propose a data augmentation approach using conditional variational auto-encoder to synthesize new failing test cases for FL. The experimental results show that our approach significantly improves six state-of-the-art fault localization techniques.
Ya ZENG Li WAN Qiuhong LUO Mao CHEN
Traditional pipeline methods for task-oriented dialogue systems are designed individually and expensively. Existing memory augmented end-to-end methods directly map the inputs to outputs and achieve promising results. However, the most existing end-to-end solutions store the dialogue history and knowledge base (KB) information in the same memory and represent KB information in the form of KB triples, making the memory reader's reasoning on the memory more difficult, which makes the system difficult to retrieve the correct information from the memory to generate a response. Some methods introduce many manual annotations to strengthen reasoning. To reduce the use of manual annotations, while strengthening reasoning, we propose a hierarchical memory model (HM2Seq) for task-oriented systems. HM2Seq uses a hierarchical memory to separate the dialogue history and KB information into two memories and stores KB in KB rows, then we use memory rows pointer combined with an entity decoder to perform hierarchical reasoning over memory. The experimental results on two publicly available task-oriented dialogue datasets confirm our hypothesis and show the outstanding performance of our HM2Seq by outperforming the baselines.
Shi Ping CAI Zhi HU Chang An ZHAO
The final exponentiation affects the efficiency of pairing computations especially on pairing-friendly curves with high embedding degree. We propose an efficient method for computing the hard part of the final exponentiation on the KSS18 curve at the 192-bit security level. Implementations indicate that the computation of the final exponentiation is 8.74% faster than the previously fastest result.
Kazuto FUKUCHI Chia-Mu YU Jun SAKUMA
We investigate a problem of finding the minimum, in which each user has a real value, and we want to estimate the minimum of these values under the local differential privacy constraint. We reveal that this problem is fundamentally difficult, and we cannot construct a consistent mechanism in the worst case. Instead of considering the worst case, we aim to construct a private mechanism whose error rate is adaptive to the easiness of estimation of the minimum. As a measure of easiness, we introduce a parameter α that characterizes the fatness of the minimum-side tail of the user data distribution. As a result, we reveal that the mechanism can achieve O((ln6N/ε2N)1/2α) error without knowledge of α and the error rate is near-optimal in the sense that any mechanism incurs Ω((1/ε2N)1/2α) error. Furthermore, we demonstrate that our mechanism outperforms a naive mechanism by empirical evaluations on synthetic datasets. Also, we conducted experiments on the MovieLens dataset and a purchase history dataset and demonstrate that our algorithm achieves Õ((1/N)1/2α) error adaptively to α.
Zhaoqi LI Ta LI Qingwei ZHAO Pengyuan ZHANG
Query-by-example spoken term detection (QbE-STD) is a task of using speech queries to match utterances, and the acoustic word embedding (AWE) method of generating fixed-length representations for speech segments has shown high performance and efficiency in recent work. We propose an AWE training method using a label-adversarial network to reduce the interference information learned during AWE training. Experiments demonstrate that our method achieves significant improvements on multilingual and zero-resource test sets.
A new software based in-situ training (SBIST) method to achieve high accuracies is proposed for binarized neural networks inference accelerator chips in which measured offsets in sense amplifiers (activation binarizers) are transformed into biases in the training software. To expedite this individual training, the initial values for the weights are taken from results of a common forming training process which is conducted in advance by using the offset fluctuation distribution averaged over the fabrication line. SPICE simulation inference results for the accelerator predict that the accuracy recovers to higher than 90% even when the amplifier offset is as large as 40mV only after a few epochs of the individual training.
Naoki HIRAKURA Masaki AIDA Konosuke KAWASHIMA
While social media is now used by many people and plays a role in distributing information, it has recently created an unexpected problem: the actual shrinkage of information sources. This is mainly due to the ease of connecting people with similar opinions and the recommendation system. Biased information distribution promotes polarization that divides people into multiple groups with opposing views. Also, people may receive only the seemingly positive information that they prefer, or may trigger them into holding onto their opinions more strongly when they encounter opposing views. This, combined with the characteristics of social media, is accelerating the polarization of opinions and eventually social division. In this paper, we propose a model of opinion formation on social media to simulate polarization. While based on the idea that opinion neutrality is only relative, this model provides new techniques for dealing with polarization.
Chongzheng HAO Xiaoyu DANG Sai LI Chenghua WANG
This paper presents a deep neural network (DNN) based symbol detection and modulation classification detector (SDMCD) for mixed blind signals detection. Unlike conventional methods that employ symbol detection after modulation classification, the proposed SDMCD can perform symbol recovery and modulation identification simultaneously. A cumulant and moment feature vector is presented in conjunction with a low complexity sparse autoencoder architecture to complete mixed signals detection. Numerical results show that SDMCD scheme has remarkable symbol error rate performance and modulation classification accuracy for various modulation formats in AWGN and Rayleigh fading channels. Furthermore, the proposed detector has robust performance under the impact of frequency and phase offsets.
Phong X. NGUYEN Hung Q. CAO Khang V. T. NGUYEN Hung NGUYEN Takehisa YAIRI
In recent years, there has been an increasing trend of applying artificial intelligence in many different fields, which has a profound and direct impact on human life. Consequently, this raises the need to understand the principles of model making predictions. Since most current high-precision models are black boxes, neither the AI scientist nor the end-user profoundly understands what is happening inside these models. Therefore, many algorithms are studied to explain AI models, especially those in the image classification problem in computer vision such as LIME, CAM, GradCAM. However, these algorithms still have limitations, such as LIME's long execution time and CAM's confusing interpretation of concreteness and clarity. Therefore, in this paper, we will propose a new method called Segmentation - Class Activation Mapping (SeCAM)/ This method combines the advantages of these algorithms above while at simultaneously overcoming their disadvantages. We tested this algorithm with various models, including ResNet50, InceptionV3, and VGG16 from ImageNet Large Scale Visual Recognition Challenge (ILSVRC) data set. Outstanding results were achieved when the algorithm has met all the requirements for a specific explanation in a remarkably short space of time.
Mamoru OKUMURA Keisuke ASANO Takumi ABE Eiji OKAMOTO Tetsuya YAMAMOTO
In recent years, there has been significant interest in information-theoretic security techniques that encrypt physical layer signals. We have proposed chaos modulation, which has both physical layer security and channel coding gain, as one such technique. In the chaos modulation method, the channel coding gain can be increased using a turbo mechanism that exchanges the log-likelihood ratio (LLR) with an external concatenated code using the max-log approximation. However, chaos modulation, which is a type of Gaussian modulation, does not use fixed mapping, and the distance between signal points is not constant; therefore, the accuracy of the max-log approximated LLR degrades under poor channel conditions. As a result, conventional methods suffer from performance degradation owing to error propagation in turbo decoding. Therefore, in this paper, we propose a new LLR clipping method that can be optimally applied to chaos modulation by limiting the confidence level of LLR and suppressing error propagation. For effective clipping on chaos modulation that does not have fixed mappings, the average confidence value is obtained from the extrinsic LLR calculated from the demodulator and decoder, and clipping is performed based on this value, either in the demodulator or the decoder. Numerical results indicated that the proposed method achieves the same performance as the one using the exact LLR, which requires complicated calculations. Furthermore, the security feature of the proposed system is evaluated, and we observe that sufficient security is provided.
Shin-ichi NAKAYAMA Shigeru MASUYAMA
Given a graph G=(V, E), where V and E are vertex and edge sets of G, and a subset VNT of vertices called a non-terminal set, a spanning tree with a non-terminal set VNT, denoted by STNT, is a connected and acyclic spanning subgraph of G that contains all vertices of V where each vertex in a non-terminal set is not a leaf. On general graphs, the problem of finding an STNT of G is known to be NP-hard. In this paper, we show that if G is a circular-arc graph then finding an STNT of G is polynomially solvable with respect to the number of vertices.
Chenchen MENG Jun WANG Chengzhi DENG Yuanyun WANG Shengqian WANG
Feature representation is a key component of most visual tracking algorithms. It is difficult to deal with complex appearance changes with low-level hand-crafted features due to weak representation capacities of such features. In this paper, we propose a novel tracking algorithm through combining a joint dictionary pair learning with convolutional neural networks (CNN). We utilize CNN model that is trained on ImageNet-Vid to extract target features. The CNN includes three convolutional layers and two fully connected layers. A dictionary pair learning follows the second fully connected layer. The joint dictionary pair is learned upon extracted deep features by the trained CNN model. The temporal variations of target appearances are learned in the dictionary learning. We use the learned dictionaries to encode target candidates. A linear combination of atoms in the learned dictionary is used to represent target candidates. Extensive experimental evaluations on OTB2015 demonstrate the superior performances against SOTA trackers.