Handwriting difficulties (HWDs) in children have adverse effects on their confidence and academic progress. Detecting HWDs is the first crucial step toward clinical or teaching intervention for children with HWDs. To date, how to automatically detect HWDs is still a challenge, although digitizing tablets have provided an opportunity to automatically collect handwriting process information. Especially, to our best knowledge, there is no exploration into the potential of combining machine learning algorithms and the handwriting process information to automatically detect Chinese HWDs in children. To bridge the gap, we first conducted an experiment to collect sample data and then compared the performance of five commonly used classification algorithms (Decision tree, Support Vector Machine (SVM), Artificial Neural Network, Naïve Bayesian and k-Nearest Neighbor) in detecting HWDs. The results showed that: (1) only a small proportion (13%) of children had Chinese HWDs and each classification model on the imbalanced dataset (39 children at risk of HWDs versus 261 typical children) produced the results that were better than random guesses, indicating the possibility of using classification algorithms to detect Chinese HWDs; (2) the SVM model had the best performance in detecting Chinese HWDs among the five classification models; and (3) the performance of the SVM model, especially its sensitivity, could be significantly improved by employing the Synthetic Minority Oversampling Technique to handle the class-imbalanced data. This study gains new insights into which handwriting features are predictive of Chinese HWDs in children and proposes a method that can help the clinical and educational professionals to automatically detect children at risk of Chinese HWDs.
Jinhua DU Deng YAI Yuntian XUE Quanwei LIU
Dual-rotor machine (DRM) is a multiple input and output electromechanical device with two electrical and two mechanical ports which make it an optimal transmission system for hybrid electric vehicles. In attempt to boost its performance and efficiency, this work presents a dual-rotor permanent magnet (DR-PM) machine system used for continuously variable transmission (CVT) in HEVs. The proposed DR-PM machine is analyzed, and modeled in consideration of vehicle driving requirements. Considering energy conversion modes and torque transfer modes, operation conditions of the DR-PM machine system used for CVT are illustrated in detail. Integrated control model of the system is carried out, besides, intelligent speed ratio control strategy is designed by analyzing the dynamic coupling modes upon the integrated models to satisfy the performance requirements, reasonable energy-split between machine and engine, and optimal fuel economy. Experimental results confirm the validity of the mathematical model of the DR-PM machine system in the application of CVT, and the effectiveness of the intelligent speed ratio control strategy.
Yasser MOHAMMAD Kazunori MATSUMOTO Keiichiro HOASHI
Activity recognition from sensors is a classification problem over time-series data. Some research in the area utilize time and frequency domain handcrafted features that differ between datasets. Another categorically different approach is to use deep learning methods for feature learning. This paper explores a middle ground in which an off-the-shelf feature extractor is used to generate a large number of candidate time-domain features followed by a feature selector that was designed to reduce the bias toward specific classification techniques. Moreover, this paper advocates the use of features that are mostly insensitive to sensor orientation and show their applicability to the activity recognition problem. The proposed approach is evaluated using six different publicly available datasets collected under various conditions using different experimental protocols and shows comparable or higher accuracy than state-of-the-art methods on most datasets but usually using an order of magnitude fewer features.
Since cyber attacks such as cyberterrorism against Industrial Control Systems (ICSs) and cyber espionage against companies managing them have increased, the techniques to detect anomalies in early stages are required. To achieve the purpose, several studies have developed anomaly detection methods for ICSs. In particular, some techniques using packet flow regularity in industrial control networks have achieved high-accuracy detection of attacks disrupting the regularity, i.e. normal behaviour, of ICSs. However, these methods cannot identify scanning attacks employed in cyber espionage because the probing packets assimilate into a number of normal ones. For example, the malware called Havex is customised to clandestinely acquire information from targeting ICSs using general request packets. The techniques to detect such scanning attacks using widespread packets await further investigation. Therefore, the goal of this study was to examine high performance methods to identify anomalies even if elaborate packets to avoid alert systems were employed for attacks against industrial control networks. In this paper, a novel detection model for anomalous packets concealing behind normal traffic in industrial control networks was proposed. For the proposal of the sophisticated detection method, we took particular note of packet flow regularity and employed the Markov-chain model to detect anomalies. Moreover, we regarded not only original packets but similar ones to them as normal packets to reduce false alerts because it was indicated that an anomaly detection model using the Markov-chain suffers from the ample false positives affected by a number of normal, irregular packets, namely noise. To calculate the similarity between packets based on the packet flow regularity, a vector representation tool called word2vec was employed. Whilst word2vec is utilised for the culculation of word similarity in natural language processing tasks, we applied the technique to packets in ICSs to calculate packet similarity. As a result, the Markov-chain with word2vec model identified scanning packets assimulating into normal packets in higher performance than the conventional Markov-chain model. In conclusion, employing both packet flow regularity and packet similarity in industrial control networks contributes to improving the performance of anomaly detection in ICSs.
Tie HONG Yuan Wei LI Zhi Ying WANG
Head action recognition, as a specific problem in action recognition, has been studied in this paper. Different from most existing researches, our head action recognition problem is specifically defined for the requirement of some practical applications. Based on our definition, we build a corresponding head action dataset which contains many challenging cases. For action recognition, we proposed a real-time head action recognition framework based on HOF and ELM. The framework consists of face detection based ROI determination, HOF feature extraction in ROI, and ELM based action prediction. Experiments show that our method achieves good accuracy and is efficient enough for practical applications.
Kento HASEGAWA Masao YANAGISAWA Nozomu TOGAWA
Recently, it has been reported that malicious third-party IC vendors often insert hardware Trojans into their products. Especially in IC design step, malicious third-party vendors can easily insert hardware Trojans in their products and thus we have to detect them efficiently. In this paper, we propose a machine-learning-based hardware-Trojan detection method for gate-level netlists using multi-layer neural networks. First, we extract 11 Trojan-net feature values for each net in a netlist. After that, we classify the nets in an unknown netlist into a set of Trojan nets and that of normal nets using multi-layer neural networks. By experimentally optimizing the structure of multi-layer neural networks, we can obtain an average of 84.8% true positive rate and an average of 70.1% true negative rate while we can obtain 100% true positive rate in some of the benchmarks, which outperforms the existing methods in most of the cases.
Kehai CHEN Tiejun ZHAO Muyun YANG
Learning semantic representation for translation context is beneficial to statistical machine translation (SMT). Previous efforts have focused on implicitly encoding syntactic and semantic knowledge in translation context by neural networks, which are weak in capturing explicit structural syntax information. In this paper, we propose a new neural network with a tree-based convolutional architecture to explicitly learn structural syntax information in translation context, thus improving translation prediction. Specifically, we first convert parallel sentences with source parse trees into syntax-based linear sequences based on a minimum syntax subtree algorithm, and then define a tree-based convolutional network over the linear sequences to learn syntax-based context representation and translation prediction jointly. To verify the effectiveness, the proposed model is integrated into phrase-based SMT. Experiments on large-scale Chinese-to-English and German-to-English translation tasks show that the proposed approach can achieve a substantial and significant improvement over several baseline systems.
Makoto TAKITA Masanori HIROTOMO Masakatu MORII
The network load is increasing due to the spread of content distribution services. Caching is known as a technique to reduce a peak network load by prefetching popular contents into memories of users. Coded caching is a new caching approach based on a carefully designed content placement in order to create coded multicasting opportunities. Recent works have discussed single-layer caching systems, but many networks consist of multiple layers of cache. In this paper, we discuss a coded caching problem for a hierarchical network that has a different number of layers of cache. The network has users who connect to an origin server via a mirror server and users who directly connect to the origin server. We provide lower bounds of the rates for this problem setting based on the cut-set bound. In addition, we propose three basic coded caching schemes and characterize these schemes. Also, we propose a new coded caching scheme by combining two basic schemes and provide achievable rates of the combination coded caching scheme. Finally, we show that the proposed combination scheme demonstrates a good performance by a numerical result.
Tomohiko UYEMATSU Tetsunao MATSUTA
This paper considers a joint channel coding and random number generation from the channel output. Specifically, we want to transmit a message to a receiver reliably and at the same time the receiver extracts pure random bits independent of the channel input. We call this problem as the joint channel coding and intrinsic randomness problem. For general channels, we clarify the trade-off between the coding rate and the random bit rate extracted from the channel output by using the achievable rate region, where both the probability of decoding error and the approximation error of random bits asymptotically vanish. We also reveal the achievable rate regions for stationary memoryless channels, additive channels, symmetric channels, and mixed channels.
In 1973, Arimoto proved the strong converse theorem for the discrete memoryless channels stating that when transmission rate R is above channel capacity C, the error probability of decoding goes to one as the block length n of code word tends to infinity. He proved the theorem by deriving the exponent function of error probability of correct decoding that is positive if and only if R > C. Subsequently, in 1979, Dueck and Körner determined the optimal exponent of correct decoding. Recently the author determined the optimal exponent on the correct probability of decoding have the form similar to that of Dueck and Körner determined. In this paper we give a rigorous proof of the equivalence of the above exponet function of Dueck and Körner to a exponent function which can be regarded as an extention of Arimoto's bound to the case with the cost constraint on the channel input.
Yusaku HAYAMIZU Miki YAMAMOTO Elisha ROSENSWEIG James F. KUROSE
In-network guidance to off-path cache, Breadcrumbs, has been proposed for cache network. It guides content requests to off-path cached contents by using the latest content download direction pointer, breadcrumbs. In Breadcrumbs, breadcrumb pointer is overwritten when a new content download of the corresponding content passes through a router. There is a possibility that slightly old guidance information for popular contents might lead to better cached content than the latest one. In this paper, we propose a new in-network guidance, Multiple-Breadcrumbs, which holds old breadcrumbs even with the latest breadcrumb pointer generated with a new content download. We focus on its content search capability and propose Throughput Sensitive selection that selects the content source giving the best estimated throughput. Our performance evaluation gives interesting results that our proposed Multiple Breadcrumbs with Throughput Sensitive selection improves not only throughput for popular contents but also for unpopular contents.
Reona SUGIYAMA Quang-Thang DUONG Minoru OKADA
Optimal loads and maximum achievable efficiency for multiple-receiver inductive power transfer (IPT) system have been formulated by theoretical studies in literatures. This paper presents extended analysis on system behavior at optimal load condition and extensive S-parameter evaluation to validate the formulas. Our results confirm that at the optimal load condition, the system is in a resonance state; the impact of cross-coupling among receivers is completely mitigated; and the efficiency reaches its maximum expressed by an efficiency angle tangent, in an manner analogous to the well-known kQ-theory for single-receiver IPT. Our contributions do not lie in practical applications of multiple-receiver IPT but in establishing principles for design and benchmarking the system.
A fusion framework between CNN and RNN is proposed dedicatedly for air-writing recognition. By modeling the air-writing using both spatial and temporal features, the proposed network can learn more information than existing techniques. Performance of the proposed network is evaluated by using the alphabet and numeric datasets in the public database namely the 6DMG. Average accuracy of the proposed fusion network outperforms other techniques, i.e. 99.25% and 99.83% are observed in the alphabet gesture and the numeric gesture, respectively. Simplified structure of RNN is also proposed, which can attain about two folds speed-up of ordinary BLSTM network. It is also confirmed that only the distance between consecutive sampling points is enough to attain high recognition performance.
Yong JIN Masahiko TOMOISHI Satoshi MATSUURA Yoshiaki KITAGUCHI
Data breach and data destruction attack have become the critical security threats for the ICT (Information and Communication Technology) infrastructure. Both the Internet service providers and users are suffering from the cyber threats especially those to confidential data and private information. The requirements of human social activities make people move carrying confidential data and data breach always happens during the transportation. The Internet connectivity and cryptographic technology have made the usage of confidential data much secure. However, even with the high deployment rate of the Internet infrastructure, the concerns for lack of the Internet connectivity make people carry data with their mobile devices. In this paper, we describe the main patterns of data breach occur on mobile devices and propose a secure in-depth file system concealed by GPS-based mounting authentication to mitigate data breach on mobile devices. In the proposed in-depth file system, data can be stored based on the level of credential with corresponding authentication policy and the mounting operation will be only successful on designated locations. We implemented a prototype system using Veracrypt and Perl language and confirmed that the in-depth file system worked exactly as we expected by evaluations on two locations. The contribution of this paper includes the clarification that GPS-based mounting authentication for a file system can reduce the risk of data breach for mobile devices and a realization of prototype system.
Ghulam HUSSAIN Kamran JAVED Jundong CHO Juneho YI
Automatic monitoring of food intake in free living conditions is still an open problem to solve. This paper presents a novel necklace-type wearable system embedded with a piezoelectric sensor to monitor ingestive behavior by detecting skin motion from the lower trachea. Detected events are incorporated for food classification. Unlike the previous state-of-the-art piezoelectric sensor based system that employs spectrogram features, we have tried to fully exploit time-domain based signals for optimal features. Through numerous evaluations on the length of a frame, we have found the best performance with a frame length of 70 samples (3.5 seconds). This demonstrates that the chewing sequence carries important information for food classification. Experimental results show the validity of the proposed algorithm for food intake detection and food classification in real-life scenarios. Our system yields an accuracy of 89.2% for food intake detection and 80.3% for food classification over 17 food categories. Additionally, our system is based on a smartphone app, which helps users live healthy by providing them with real-time feedback about their ingested food episodes and types.
Practical deep neural networks have a number of weight parameters, and the dynamic fixed-point formats have been used to represent them efficiently. The dynamic fixed-point representations share an scaling factor among a group of numbers, and the weights in a layer have been formed into such a group. In this paper, we first explore a design space for dynamic fixed-point neuromorphic computing systems and show that it is indispensable to have a small group size in neuromorphic architectures, because it is appropriate to group the weights associated with a neuron into a group. We then presents a dynamic fixed-point representation designed for neuromorphic computing systems. Our experimental results show that the proposed representation reduces the required weight bitwidth by about 4 bits compared to the conventional fixed-point format.
Hyun KWON Yongchul KIM Ki-Woong PARK Hyunsoo YOON Daeseon CHOI
Deep neural networks (DNNs) are widely used in many applications such as image, voice, and pattern recognition. However, it has recently been shown that a DNN can be vulnerable to a small distortion in images that humans cannot distinguish. This type of attack is known as an adversarial example and is a significant threat to deep learning systems. The unknown-target-oriented generalized adversarial example that can deceive most DNN classifiers is even more threatening. We propose a generalized adversarial example attack method that can effectively attack unknown classifiers by using a hierarchical ensemble method. Our proposed scheme creates advanced ensemble adversarial examples to achieve reasonable attack success rates for unknown classifiers. Our experiment results show that the proposed method can achieve attack success rates for an unknown classifier of up to 9.25% and 18.94% higher on MNIST data and 4.1% and 13% higher on CIFAR10 data compared with the previous ensemble method and the conventional baseline method, respectively.
Koichi MITSUNARI Jaehoon YU Takao ONOYE Masanori HASHIMOTO
Visual object detection on embedded systems involves a multi-objective optimization problem in the presence of trade-offs between power consumption, processing performance, and detection accuracy. For a new Pareto solution with high processing performance and low power consumption, this paper proposes a hardware architecture for decision tree ensemble using multiple channels of features. For efficient detection, the proposed architecture utilizes the dimensionality of feature channels in addition to parallelism in image space and adopts task scheduling to attain random memory access without conflict. Evaluation results show that an FPGA implementation of the proposed architecture with an aggregated channel features pedestrian detector can process 229 million samples per second at 100MHz operation frequency while it requires a relatively small amount of resources. Consequently, the proposed architecture achieves 350fps processing performance for 1080P Full HD images and outperforms conventional object detection hardware architectures developed for embedded systems.
Maoxi LI Qingyu XIANG Zhiming CHEN Mingwen WANG
The-state-of-the-art neural quality estimation (QE) of machine translation model consists of two sub-networks that are tuned separately, a bidirectional recurrent neural network (RNN) encoder-decoder trained for neural machine translation, called the predictor, and an RNN trained for sentence-level QE tasks, called the estimator. We propose to combine the two sub-networks into a whole neural network, called the unified neural network. When training, the bidirectional RNN encoder-decoder are initialized and pre-trained with the bilingual parallel corpus, and then, the networks are trained jointly to minimize the mean absolute error over the QE training samples. Compared with the predictor and estimator approach, the use of a unified neural network helps to train the parameters of the neural networks that are more suitable for the QE task. Experimental results on the benchmark data set of the WMT17 sentence-level QE shared task show that the proposed unified neural network approach consistently outperforms the predictor and estimator approach and significantly outperforms the other baseline QE approaches.
Ken-ichiro MORIDOMI Kohei HATANO Eiji TAKIMOTO
We prove generalization error bounds of classes of low-rank matrices with some norm constraints for collaborative filtering tasks. Our bounds are tighter, compared to known bounds using rank or the related quantity only, by taking the additional L1 and L∞ constraints into account. Also, we show that our bounds on the Rademacher complexity of the classes are optimal.