Lihan TONG Weijia LI Qingxia YANG Liyuan CHEN Peng CHEN
Yinan YANG
Myung-Hyun KIM Seungkwang LEE
Shuoyan LIU Chao LI Yuxin LIU Yanqiu WANG
Takumi INABA Takatsugu ONO Koji INOUE Satoshi KAWAKAMI
Martin LUKAC Saadat NURSULTAN Georgiy KRYLOV Oliver KESZOCZE Abilmansur RAKHMETTULAYEV Michitaka KAMEYAMA
Zheqing ZHANG Hao ZHOU Chuan LI Weiwei JIANG
Liu ZHANG Zilong WANG Yindong CHEN
Wenxia Bao An Lin Hua Huang Xianjun Yang Hemu Chen
Fengshan ZHAO Qin LIU Takeshi IKENAGA
Haruhiko KAIYA Shinpei OGATA Shinpei HAYASHI
Jiakai LI Jianyong DUAN Hao WANG Li HE Qing ZHANG
Yuxin HUANG Yuanlin YANG Enchang ZHU Yin LIANG Yantuan XIAN
Naohito MATSUMOTO Kazuhiro KURITA Masashi KIYOMI
Na XING Lu LI Ye ZHANG Shiyi YANG
Zhe Wang Zhe-Ming Lu Hao Luo Yang-Ming Zheng
Rina TAGAMI Hiroki KOBAYASHI Shuichi AKIZUKI Manabu HASHIMOTO
Tomohiro KOBAYASHI Tomomi MATSUI
Shin-ichi NAKANO
Hongzhi XU Binlian ZHANG
Weizhi WANG Lei XIA Zhuo ZHANG Xiankai MENG
Yuka KO Katsuhito SUDOH Sakriani SAKTI Satoshi NAKAMURA
Rinka KAWANO Masaki KAWAMURA
Zhishuo ZHANG Chengxiang TAN Xueyan ZHAO Min YANG
Peng WANG Guifen CHEN Zhiyao SUN
Zeyuan JU Zhipeng LIU Yu GAO Haotian LI Qianhang DU Kota YOSHIKAWA Shangce GAO
Ji WU Ruoxi YU Kazuteru NAMBA
Hao WANG Yao Ma Jianyong Duan Li HE Xin Li
Shijie WANG Xuejiao HU Sheng LIU Ming LI Yang LI Sidan DU
Arata KANEKO Htoo Htoo Sandi KYAW Kunihiro FUJIYOSHI Keiichi KANEKO
Qi LIU Bo WANG Shihan TAN Shurong ZOU Wenyi GE
HanYu Zhang Tomoji Kishi
Shinobu NAGAYAMA Tsutomu SASAO Jon T. BUTLER
Yoon Hak KIM
Takashi HIRAYAMA Rin SUZUKI Katsuhisa YAMANAKA Yasuaki NISHITANI
Yosuke IIJIMA Atsunori OKADA Yasushi YUMINAKA
Batnasan Luvaanjalba Elaine Yi-Ling Wu
KuanChao CHU Satoshi YAMAZAKI Hideki NAKAYAMA
Shenglei LI Haoran LUO Tengfei SHAO Reiko HISHIYAMA
Yasushi YUMINAKA Kazuharu NAKAJIMA Yosuke IIJIMA
Chunbo Liu Liyin Wang Zhikai Zhang Chunmiao Xiang Zhaojun Gu Zhi Wang Shuang Wang
Jia-ji JIANG Hai-bin WAN Hong-min SUN Tuan-fa QIN Zheng-qiang WANG
Yuhao LIU Zhenzhong CHU Lifei WEI
Ken ASANO Masanori NATSUI Takahiro HANYU
Shuto HASEGAWA Koichiro ENOMOTO Taeko MIZUTANI Yuri OKANO Takenori TANAKA Osamu SAKAI
Zhewei XU Mizuho IWAIHARA
Takao WAHO Akihisa KOYAMA Hitoshi HAYASHI
Taisei SAITO Kota ANDO Tetsuya ASAI
Shiyu YANG Tetsuya KANDA Daniel M. GERMAN Yoshiki HIGO
Tsutomu SASAO
Jiyeon LEE
Koichi MORIYAMA Akira OTSUKA
Hongliang FU Qianqian LI Huawei TAO Chunhua ZHU Yue XIE Ruxue GUO
Gao WANG Gaoli WANG Siwei SUN
Hua HUANG Yiwen SHAN Chuan LI Zhi WANG
Zhi LIU Heng WANG Yuan LI Hongyun LU Hongyuan JING Mengmeng ZHANG
Tomoyasu NAKANO Masataka GOTO
Hyebong CHOI Joel SHIN Jeongho KIM Samuel YOON Hyeonmin PARK Hyejin CHO Jiyoung JUNG
Xianglong LI Yuan LI Jieyuan ZHANG Xinhai XU Donghong LIU
Haoran LUO Tengfei SHAO Shenglei LI Reiko HISHIYAMA
Chang SUN Yitong LIU Hongwen YANG
Ji XI Yue XIE Pengxu JIANG Wei JIANG
Ming PAN
The increasing amount of fake news is a growing problem that will progressively worsen in our interconnected world. Machine learning, particularly deep learning, is being used to detect misinformation; however, the models employed are essentially black boxes, and thus are uninterpretable. This paper presents an overview of explainable fake news detection models. Specifically, we first review the existing models, datasets, evaluation techniques, and visualization processes. Subsequently, possible improvements in this field are identified and discussed.
Duc A. HOANG Akira SUZUKI Tsuyoshi YAGITA
A vertex subset I of a graph G is called a k-path vertex cover if every path on k vertices in G contains at least one vertex from I. The K-PATH VERTEX COVER RECONFIGURATION (K-PVCR) problem asks if one can transform one k-path vertex cover into another via a sequence of k-path vertex covers where each intermediate member is obtained from its predecessor by applying a given reconfiguration rule exactly once. We investigate the computational complexity of K-PVCR from the viewpoint of graph classes under the well-known reconfiguration rules: TS, TJ, and TAR. The problem for k=2, known as the VERTEX COVER RECONFIGURATION (VCR) problem, has been well-studied in the literature. We show that certain known hardness results for VCR on different graph classes can be extended for K-PVCR. In particular, we prove a complexity dichotomy for K-PVCR on general graphs: on those whose maximum degree is three (and even planar), the problem is PSPACE-complete, while on those whose maximum degree is two (i.e., paths and cycles), the problem can be solved in polynomial time. Additionally, we also design polynomial-time algorithms for K-PVCR on trees under each of TJ and TAR. Moreover, on paths, cycles, and trees, we describe how one can construct a reconfiguration sequence between two given k-path vertex covers in a yes-instance. In particular, on paths, our constructed reconfiguration sequence is shortest.
Dehua LIANG Jun SHIOMI Noriyuki MIURA Masanori HASHIMOTO Hiromitsu AWANO
Reservoir computing (RC) is an attractive alternative to machine learning models owing to its computationally inexpensive training process and simplicity. In this work, we propose EnsembleBloomCA, which utilizes cellular automata (CA) and an ensemble Bloom filter to organize an RC system. In contrast to most existing RC systems, EnsembleBloomCA eliminates all floating-point calculation and integer multiplication. EnsembleBloomCA adopts CA as the reservoir in the RC system because it can be implemented using only binary operations and is thus energy efficient. The rich pattern dynamics created by CA can map the original input into a high-dimensional space and provide more features for the classifier. Utilizing an ensemble Bloom filter as the classifier, the features provided by the reservoir can be effectively memorized. Our experiment revealed that applying the ensemble mechanism to the Bloom filter resulted in a significant reduction in memory cost during the inference phase. In comparison with Bloom WiSARD, one of the state-of-the-art reference work, the EnsembleBloomCA model achieves a 43× reduction in memory cost while maintaining the same accuracy. Our hardware implementation also demonstrated that EnsembleBloomCA achieved over 23× and 8.5× reductions in area and power, respectively.
Manufacturers are coping with increasing pressures in quality, cost and efficiency as more and more industries are moving from traditional setup to industry 4.0 based digitally transformed setup due to its numerous playbacks. Within the manufacturing domain organizational structures and processes are complex, therefore adopting industry 4.0 and finding an optimized re-engineered business process is difficult without using a systematic methodology. Authors have developed Business Process Re-engineering (BPR) and Business Process Optimization (BPO) methods but no consolidated methodology have been seen in the literature that is based on industry 4.0 and incorporates both the BPR and BPO. We have presented a consolidated and systematic re-engineering and optimization framework for a manufacturing industry setup. The proposed framework performs Evolutionary Multi-Objective Combinatorial Optimization using Multi-Objective Genetic Algorithm (MOGA). An example process from an aircraft manufacturing factory has been optimized and re-engineered with available set of technologies from industry 4.0 based on the criteria of lower cost, reduced processing time and reduced error rate. At the end to validate the proposed framework Business Process Model and Notation (BPMN) is used for simulations and perform comparison between AS-IS and TO-BE processes as it is widely used standard for business process specification. The proposed framework will be used in converting an industry from traditional setup to industry 4.0 resulting in cost reduction, increased performance and quality.
Jinho CHOI Jaehan KIM Minkyoo SONG Hanna KIM Nahyeon PARK Minjae SEO Youngjin JIN Seungwon SHIN
Cryptocurrency abuse has become a critical problem. Due to the anonymous nature of cryptocurrency, criminals commonly adopt cryptocurrency for trading drugs and deceiving people without revealing their identities. Despite its significance and severity, only few works have studied how cryptocurrency has been abused in the real world, and they only provide some limited measurement results. Thus, to provide a more in-depth understanding on the cryptocurrency abuse cases, we present a large-scale analysis on various Bitcoin abuse types using 200,507 real-world reports collected by victims from 214 countries. We scrutinize observable abuse trends, which are closely related to real-world incidents, to understand the causality of the abuses. Furthermore, we investigate the semantics of various cryptocurrency abuse types to show that several abuse types overlap in meaning and to provide valuable insight into the public dataset. In addition, we delve into abuse channels to identify which widely-known platforms can be maliciously deployed by abusers following the COVID-19 pandemic outbreak. Consequently, we demonstrate the polarization property of Bitcoin addresses practically utilized on transactions, and confirm the possible usage of public report data for providing clues to track cyber threats. We expect that this research on Bitcoin abuse can empirically reach victims more effectively than cybercrime, which is subject to professional investigation.
Thi Thu Thao KHONG Takashi NAKADA Yasuhiko NAKASHIMA
We introduce a hybrid Bayesian-convolutional neural network (hyBCNN) for improving the robustness against adversarial attacks and decreasing the computation time in the Bayesian inference phase. Our hyBCNN models are built from a part of BNN and CNN. Based on pre-trained CNNs, we only replace convolutional layers and activation function of the initial stage of CNNs with our Bayesian convolutional (BC) and Bayesian activation (BA) layers as a term of transfer learning. We keep the remainder of CNNs unchanged. We adopt the Bayes without Bayesian Learning (BwoBL) algorithm for hyBCNN networks to execute Bayesian inference towards adversarial robustness. Our proposal outperforms adversarial training and robust activation function, which are currently the outstanding defense methods of CNNs in the resistance to adversarial attacks such as PGD and C&W. Moreover, the proposed architecture with BwoBL can easily integrate into any pre-trained CNN, especially in scaling networks, e.g., ResNet and EfficientNet, with better performance on large-scale datasets. In particular, under l∞ norm PGD attack of pixel perturbation ε=4/255 with 100 iterations on ImageNet, our best hyBCNN EfficientNet reaches 93.92% top-5 accuracy without additional training.
The shared last level cache (SLLC) in tile chip multiprocessors (TCMP) provides a low off-chip miss rate, but it causes a long on-chip access latency. In the two-level cache hierarchy, data replication stores replicas of L1 victims in the local LLC (L2 cache) to obtain a short local LLC access latency on the next accesses. Many data replication mechanisms have been proposed, but they do not consider both L1 victim reuse behaviors and LLC replica reception capability. They either produce many useless replicas or increase LLC pressure, which limits the improvement of system performance. In this paper, we propose a two-level cache aware adaptive data replication mechanism (TCDR), which controls replication based on both L1 victim reuse behaviors prediction and LLC replica reception capability monitoring. TCDR not only increases the accuracy of L1 replica selection, but also avoids the pressure of replication on LLC. The results show that TCDR improves the system performance with reasonable hardware overhead.
Takashi YOKOTA Kanemitsu OOTSU Shun KOJIMA
Parallel computing essentially consists of computation and communication and, in many cases, communication performance is vital. Many parallel applications use collective communications, which often dominate the performance of the parallel execution. This paper focuses on collective communication performance to speed-up the parallel execution. This paper firstly offers our experimental result that splitting a session of collective communication to small portions (slices) possibly enables efficient communication. Then, based on the results, this paper proposes a new concept cup-stacking with a genetic algorithm based methodology. The preliminary evaluation results reveal the effectiveness of the proposed method.
Kyungmin KIM Jiung SONG Jong Wook KWAK
We propose a novel synthetic-benchmarks generation model using partial time-series regression, called Partial-Regression-Integrated Generic Model (PRIGM). PRIGM abstracts the unique characteristics of the input sensor data into generic time-series data confirming the generation similarity and evaluating the correctness of the synthetic benchmarks. The experimental results obtained by the proposed model with its formula verify that PRIGM preserves the time-series characteristics of empirical data in complex time-series data within 10.4% on an average difference in terms of descriptive statistics accuracy.
In recent years, Graph Neural Networks has received enormous attention from academia for its huge potential of modeling the network traits such as macrostructure and single node attributes. However, prior mainstream works mainly focus on homogeneous network and lack the capacity to characterize the network heterogeneous property. Besides, most previous literature cannot the model latent influence link under microscope vision, making it infeasible to model the joint relation between the heterogeneity and mutual interaction within multiple relation type. In this letter, we propose a latent influence based self-attention framework to address the difficulties mentioned above. To model the heterogeneity and mutual interactions, we redesign the attention mechanism with latent influence factor on single-type relation level, which learns the importance coefficient from its adjacent neighbors under the same meta-path based patterns. To incorporate the heterogeneous meta-path in a unified dimension, we developed a novel self-attention based framework for meta-path relation fusion according to the learned meta-path coefficient. Our experimental results demonstrate that our framework not only achieves higher results than current state-of-the-art baselines, but also shows promising vision on depicting heterogeneous interactive relations under complicated network structure.
Loan default prediction has been a significant problem in the financial domain because overdue loans may incur significant losses. Machine learning methods have been introduced to solve this problem, but there are still many challenges including feature multicollinearity, imbalanced labels, and small data sample problems. To replicate the success of deep learning in many areas, an effective regularization technique named muddling label regularization is introduced in this letter, and an ensemble of feed-forward neural networks is proposed, which outperforms machine learning and deep learning baselines in a real-world dataset.
This letter presents a new framework for synchronous remote online exams. This framework proposes new monitoring of notebooks in remote locations and limited messaging only enabled between students and their instructor during online exams. This framework was evaluated by students as highly effective in minimizing cheating during online exams.
This letter presents an innovative solution for real-time interaction during online classes. Synchronous sharing enables instructors to provide real-time feedback to students. This encourages students to stay focused and feel engaged during class. Consequently, students evaluated anonymously that this solution significantly enhanced their learning experience during real-time online classes.
Yue XIE Ruiyu LIANG Xiaoyan ZHAO Zhenlin LIANG Jing DU
To alleviate the problem of the dependency on the quantity of the training sample data in speech emotion recognition, a weighted gradient pre-train algorithm for low-resource speech emotion recognition is proposed. Multiple public emotion corpora are used for pre-training to generate shared hidden layer (SHL) parameters with the generalization ability. The parameters are used to initialize the downsteam network of the recognition task for the low-resource dataset, thereby improving the recognition performance on low-resource emotion corpora. However, the emotion categories are different among the public corpora, and the number of samples varies greatly, which will increase the difficulty of joint training on multiple emotion datasets. To this end, a weighted gradient (WG) algorithm is proposed to enable the shared layer to learn the generalized representation of different datasets without affecting the priority of the emotion recognition on each corpus. Experiments show that the accuracy is improved by using CASIA, IEMOCAP, and eNTERFACE as the known datasets to pre-train the emotion models of GEMEP, and the performance could be improved further by combining WG with gradient reversal layer.
Xiaozhou CHENG Rui LI Yanjing SUN Yu ZHOU Kaiwen DONG
Visible-Infrared Person Re-identification (VI-ReID) is a challenging pedestrian retrieval task due to the huge modality discrepancy and appearance discrepancy. To address this tough task, this letter proposes a novel gray augmentation exploration (GAE) method to increase the diversity of training data and seek the best ratio of gray augmentation for learning a more focused model. Additionally, we also propose a strong all-modality center-triplet (AMCT) loss to push the features extracted from the same pedestrian more compact but those from different persons more separate. Experiments conducted on the public dataset SYSU-MM01 demonstrate the superiority of the proposed method in the VI-ReID task.