检索结果-南通市图书馆

TP-MobNet: A Two-pass Mobile Network for Low-complexity Classification of Acoustic Scene

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第11期73卷 3291-3303页

作者： Soonshin Seo Junseok Oh Eunsoo Cho Hosung Park Gyujin kim ji-hwan kim NAVER Corporation Seongnam13561Korea Department of Computer Science and Engineering Sogang UniversitySeoul04107Korea

Acoustic scene classification(ASC)is a method of recognizing and classifying environments that employ acoustic *** ASC approaches based on deep learning have been developed,with convolutional neural networks(CNNs)proving to be the most reliable and commonly utilized in ASC systems due to their suitability for constructing lightweight *** using ASC systems in the real world,model complexity and device robustness are essential *** this paper,we propose a two-pass mobile network for low-complexity classification of the acoustic scene,named *** inverse residuals and linear bottlenecks,TPMobNet is based on MobileNetV2,and following mobile blocks,coordinate attention and two-pass fusion approaches are *** log-range dependencies and precise position information in feature maps can be trained via coordinate *** capturing more diverse feature resolutions at the network’s end sides,two-pass fusions can also train ***,the model size is reduced by applying weight quantization to the trained *** adding weight quantization to the trained model,the model size is also *** TAU Urban Acoustic Scenes 2020 Mobile development set was used for all of the *** has been confirmed that the proposed model,with a model size of 219.6 kB,achieves an accuracy of 73.94%.

关键词： Acoustic scene classification low-complexity device robustness two-pass mobile network coordinate attention weight quantization

Subcarrier BD with Cooperative Communication for MIMO-NOMA System

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第9期72卷 5807-5821页

作者： Jung-In Baik ji-hwan kim Beom-Sik Shin ji-Hye Oh Hyoung-Kyu Song Department of Information and Communication Engineering and Convergence for Intelligent DroneSejong UniversitySeoul05006Korea

With the rapid evolution of Internet of things(IoT),many edge devices require simultaneous connection in 5G communication *** afford massive data of IoT devices,multiple input multiple output non-orthogonal multiple access(MIMO-NOMA)method has been considered as a promising ***,there are numerous drawbacks due to error propagation and inter-user ***,proposed scheme aims to improve the reliability of the MIMO-NOMA system with digital beamforming and intracluster cooperative multi point(CoMP)to efficiently support IoT *** the conventional MIMO-NOMA system,user entities are grouped into *** diagonalization(BD)is applied to efficiently eliminate the inter-cluster interference of the MIMO-NOMA ***,since the channel path of the data stream from a single antenna to a single cluster doesn’t hold other cluster’s data,the system can’t fully utilize the selective subcarrier channel *** indicates that there can be better channel paths for a data stream at a certain subcarrier ***,proposed scheme allocates data streams to antennas adaptively considering selective channel ***,intra-cluster CoMP method is adjusted to enhance the reliability of the system in the *** simulation results show that the proposed scheme improves BER and throughput performance compared to the conventional MIMO-NOMA system.

关键词： 5G MIMO-NOMA CoMP BD interference

Language Model Using Differentiable Neural Computer Based on Forget Gate-Based Memory Deallocation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2021年第7期68卷 537-551页

作者： Donghyun Lee Hosung Park Soonshin Seo Changmin kim Hyunsoo Son Gyujin kim ji-hwan kim Department of Computer Science and Engineering Sogang UniversitySeoul04107Korea

A differentiable neural computer(DNC)is analogous to the Von Neumann machine with a neural network controller that interacts with an external memory through an attention *** DNC’s offer a generalized method for task-specific deep learning models and have demonstrated reliability with reasoning *** this study,we apply a DNC to a language model(LM)*** LM task is one of the reasoning problems,because it can predict the next word using the previous word ***,memory deallocation is a problem in DNCs as some information unrelated to the input sequence is not allocated and remains in the external memory,which degrades ***,we propose a forget gatebased memory deallocation(FMD)method,which searches for the minimum value of elements in a forget gate-based retention *** forget gatebased retention vector indicates the retention degree of information stored in each external memory *** experiments,we applied our proposed NTM architecture to LM tasks as a task-specific example and to rescoring for speech recognition as a general-purpose *** LM tasks,we evaluated DNC using the Penn Treebank and enwik8 LM *** it does not yield SOTA results in LM tasks,the FMD method exhibits relatively improved performance compared with DNC in terms of *** the speech recognition rescoring tasks,FMD again showed a relative improvement using the LibriSpeech data in terms of word error rate.

关键词： Forget gate-based memory deallocation differentiable neural computer language model forget gate-based retention vector

Long Short-Term Memory Recurrent Neural Network-Based Acoustic Model Using Connectionist Temporal Classification on a Large-Scale Training Corpus

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

China Communications 2017年第9期14卷 23-31页

作者： Donghyun Lee Minkyu Lim Hosung Park Yoseb Kang Jeong-Sik Park Gil-jin Jang ji-hwan kim Department of Computer Science and Engineering Sogang University Department of Information and Communication Engineering Yeungnam University School of Electronics Engineering Kyungpook National University

A Long Short-Term Memory(LSTM) Recurrent Neural Network(RNN) has driven tremendous improvements on an acoustic model based on Gaussian Mixture Model(GMM). However, these models based on a hybrid method require a forced aligned Hidden Markov Model(HMM) state sequence obtained from the GMM-based acoustic model. Therefore, it requires a long computation time for training both the GMM-based acoustic model and a deep learning-based acoustic model. In order to solve this problem, an acoustic model using CTC algorithm is proposed. CTC algorithm does not require the GMM-based acoustic model because it does not use the forced aligned HMM state sequence. However, previous works on a LSTM RNN-based acoustic model using CTC used a small-scale training corpus. In this paper, the LSTM RNN-based acoustic model using CTC is trained on a large-scale training corpus and its performance is evaluated. The implemented acoustic model has a performance of 6.18% and 15.01% in terms of Word Error Rate(WER) for clean speech and noisy speech, respectively. This is similar to a performance of the acoustic model based on the hybrid method.

关键词： acoustic model connectionisttemporal classification large-scale trainingcorpus long short-term memory recurrentneural network

Hybridization of different types of exceptional points

在线全文

学校读者我要写书评

暂无评论

Photonics Research 2019年第12期7卷 1473-1478页

作者： jiNHYEOK RYU SUNJAE GWAK JAEWON kim HYEON-HYE YU ji-hwan kim ji-WON LEE CHANG-hwan YI CHIL-MIN kim Department of Emerging Materials Science DGISTDaegu 42988South Korea Institut fur Physik Technische Universitat IlmenauD-98693 IlmenauGermany Samsung Electro-Mechanics Suwon 16674South Korea

A large number of different types of second-order non-Hermitian degeneracies called exceptional points(EPs)were found in various physical systems depending on the mechanism of coupling between *** show that these EPs can be hybridized to form higher-order EPs,which preserve the original properties of the initial EPs before *** a demonstration,we hybridize chiral and supermode second-order EPs,where the former and the latter are the results of intra-disk and inter-disk mode coupling in an optical system comprised of two Mie-scale microdisks and one Rayleigh-scale *** high sensitivity of the resulting third-order EP against external perturbations in our feasible system is emphasized.

关键词： exceptional coupling system

Impact of non-Hermitian mode interaction on inter-cavity light transfer

在线全文

学校读者我要写书评

暂无评论

Photonics Research 2022年第5期10卷 1232-1237页

作者： HYEON-HYE YU SUNJAE GWAK jiNHYEOK RYU HYUNDONG kim ji-hwan kim JUNG WAN RYU CHIL-MIN kim CHANG hwan YI Department of Emerging Materials Science DGISTDaegu 42988Republic of Korea Center for Theoretical Physics of Complex Systems Institute for Basic Science(IBS)Daejeon 34126Republic of Korea

Understanding inter-site mutual mode interaction in coupled physical systems is essential to comprehend large compound systems,as this local interaction determines the successive multiple inter-site energy transfer *** the present study,we demonstrate that only the non-Hermitian coupling can correctly account for the light transfer between two coupled optical *** also reveal that the non-Hermitian coupling effect becomes crucial as the system dimension *** results provide important insight for handling general-coupled devices in the subwavelength regime.

关键词： interaction transfer coupling

Trend of Developing Aqueous Liquid and Gel Electrolytes for Sustainable,Safe,and High‑Performance Li‑Ion Batteries

在线全文

学校读者我要写书评

暂无评论

Nano-Micro Letters 2024年第1期16卷 17-34页

作者： Donghwan ji Jaeyun kim School of Chemical Engineering Sungkyunkwan University(SKKU)Suwon 16419Republic of Korea Department of Health Sciences and Technology Samsung Advanced Institute for Health Sciences&Technology(SAIHST)Sungkyunkwan University(SKKU)Suwon 16419Republic of Korea Biomedical Institute for Convergence at SKKU(BICS) Sungkyunkwan University(SKKU)Suwon 16419Republic of Korea Institute of Quantum Biophysics(IQB) Sungkyunkwan University(SKKU)Suwon 16419Republic of Korea Present Address:Department of NanoEngineering University of California San DiegoLa JollaSan DiegoCA 92093USA

Current lithium-ion batteries(LIBs)rely on organic liquid electrolytes that pose significant risks due to their flammability and *** potential for environmental pollution and explosions resulting from battery damage or fracture is a critical ***-based(aqueous)electrolytes have been receiving attention as an alternative to organic ***,a narrow electrochemicalstability window,water decomposition,and the consequent low battery operating voltage and energy density hinder the practical use of aqueous ***,developing novel aqueous electrolytes for sustainable,safe,high-performance LIBs remains *** Review first commences by summarizing the roles and requirements of electrolytes–separators and then delineates the progression of aqueous electrolytes for LIBs,encompassing aqueous liquid and gel electrolyte development trends along with detailed principles of the *** aqueous electrolytes are progressed based on strategies using superconcentrated salts,concentrated diluents,polymer additives,polymer networks,and artificial passivation layers,which are used for suppressing water decomposition and widening the electrochemical stability window of water of the *** addition,this Review discusses potential strategies for the implementation of aqueous Li-metal batteries with improved electrolyte–electrode interfaces.A comprehensive understanding of each strategy in the aqueous system will assist in the design of an aqueous electrolyte and the development of sustainable and safe high-performance batteries.

关键词： Lithium-ion battery(LIB) Aqueous electrolyte Gel electrolyte Electrochemical stability window Li dendrite

The SARS-CoV-2 spike protein induces lung cancer migration and invasion in a TLR2-dependent manner

在线全文

学校读者我要写书评

暂无评论

Cancer Communications 2024年第2期44卷 273-277页

作者： Mi-Jeong kim ji Young kim ji Hye Shin Juhee Son Yeeun Kang Soo-Kyung Jeong Duk-hwan kim Kyun-hwan kim Eunyoung Chun Ki-Young Lee Department of Immunology Samsung Biomedical Research InstituteSungkyunkwan University School of MedicineSuwonRepublic of Korea Research and Development Center CHA Vaccine InstituteSeongnam-siRepublic of Korea Department of Molecular Cell Biology Sungkyunkwan University School of MedicineSuwonRepublic of Korea Department of Precision Medicine Sungkyunkwan University School of MedicineSuwonRepublic of Korea Department of Health Sciences and Technology Samsung Advanced Institute for Health Sciences&TechnologySamsung Medical CenterSungkyunkwan UniversitySeoulRepublic of Korea

Dear editor,Severe acute respiratory syndrome coronavirus2(SARS-CoV-2)infection can lead to severe outcomes in patients with cancer[1].It has been reported that patients with lung cancers disproportionately manifest severe COVID-19 with a high rate of hospitalization and death[2].Notably,the SARS-CoV-2 Spike(S)protein can induce hyper-inflammation in both epithelial cells and macrophages through toll-like receptor(TLR)1/TLR2 or TLR2/6-dependent nuclear factor-kappaB(NF-κB)path-way[3].However,molecular and cellular evidence on whether the SARS-CoV-2 virus affects the severity of lung cancer patients through TLR1/2 or TLR2/6 signaling remains unclear.

关键词： lung TLR2 invasion

Disordered Structure and Reversible Phase Transformation from K-Birnessite to Zn-Buserite Enable High-Performance Aqueous Zinc-lon Batteries

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Energy & Environmental Materials 2024年第3期7卷 100-111页

作者： Nibagani Naresh Suyoon Eom Sang Jun Lee Su hwan Jeong ji-Won Jung Young Hwa Jung Joo-Hyung kim Department of Materials Engineering and Convergence Technology Gyeongsang National University(GNU)Jinju 52828South Korea School of Material Science and Engineering(Ceramic Engineering) Gyeongsang National University(GNU)Jinju 52828South Korea School of Materials Science and Engineering University of Ulsan(UOU)Ulsan 44776South Korea PLS-ⅡBeamline Division Pohang Accelerator Laboratory(PAL)Pohang 37673South Korea

The layeredδ-MnO_(2)(dMO)is an excellent cathode material for rechargeable aqueous zinc-ion batteries owing to its large interlayer distance(~0.7 nm),high capacity,and low cost;however,such cathodes suffer from structural degradation during the long-term cycling process,leading to capacity *** this study,a Co-doped dMO composite with reduced graphene oxide(GC-dMO)is developed using a simple cost-effective hydrothermal *** degree of disorderness increases owing to the hetero-atom doping and graphene oxide *** is demonstrated that layered dMO and GC-dMO undergo a structural transition from K-birnessite to the Zn-buserite phase upon the first discharge,which enhances the intercalation of Zn^(2+)ions,H_(2)O molecules in the layered *** GC-dMO cathode exhibits an excellent capacity of 302 mAh g^(-1)at a current density of 100 mAg^(-1)after 100 cycles as compared with the dMO cathode(159 mAhg^(-1)).The excellent electrochemical performance of the GC-dMO cathode owing to Co-doping and graphene oxide sheets enhances the interlayer gap and disorderness,and maintains structural stability,which facilitates the easy reverse intercalation and de-intercalation of Zn^(2+)ions and H_(2)O ***,GC-dMO is a promising cathode material for large-scale aqueous ZIBs.

关键词： aqueous zinc-ion batteries birnessite buserite disordered structure phase transformation