检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Ge, Meng Xu, Chenglin Wang, Longbiao Chng, Eng Siong Dang, Jianwu Li, Haizhou Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China School of Computer Science and Engineering Nanyang Technological University Singapore Japan Advanced Institute of Science and Technology Ishikawa Japan Department of Electrical and Computer Engineering National University of Singapore Singapore

Speaker extraction aims to extract the target speech signal from a multi-talker environment given a target speaker’s reference speech. We recently proposed a time-domain solution, SpEx, that avoids the phase estimation in frequency-domain approaches. Unfortunately, SpEx is not fully a time-domain solution since it performs time-domain speech encoding for speaker extraction, while taking frequency-domain speaker embedding as the reference. The size of the analysis window for time-domain and the size for frequency-domain input are also different. Such mismatch has an adverse effect on the system performance. To eliminate such mismatch, we propose a complete time-domain speaker extraction solution, that is called SpEx+. Specifically, we tie the weights of two identical speech encoder networks, one for the encoder-extractor-decoder pipeline, another as part of the speaker encoder. Experiments show that the SpEx+ achieves 0.8dB and 2.1dB SDR improvement over the state-of-the-art SpEx baseline, under different and same gender conditions on WSJ0-2mix-extr database respectively. Copyright © 2020, The Authors. All rights reserved.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

An integrated system for robust gender classification with convolutional restricted Boltzmann machine and spiking neural network

An integrated system for robust gender classification with c...

引用

IEEE Symposium Series on Computational Intelligence (SSCI)

作者： Yanli Yao Qiang Yu Longbiao Wang Jianwu Dang Tianjin Key Laboratory of Cognitive Computing and Application School of Computer Science and Technology Tianjin University Tianjin China

ISBN: (数字)9781728124858

ISBN: (纸本)9781728124865

Different from traditional artificial neural networks (ANNs), spiking neural networks (SNNs) represent and transfer information in spikes, which are considered more like human brain. SNNs contain time information, which make them more suitable for addressing time-structured speech signals. However, it still remains challenging for spiking neural network (SNN) to implement classification tasks based on speech due to the lack of a proper encoding. In this paper, an integrated spiking neural network is proposed to perform the gender classification task. The whole system consists of three functional parts, including encoding, learning and readout. As convolutional restricted Boltzmann machine (CRBM) has shown outstanding capability for unsupervised learning of auditory features, we adopt it in this paper as a feature extractor, followed by a spike-latency encoding layer that converts the feature values into spike times. Then these spikes are processed by the spiking neural networks with the tempotron learning rule. We use the TIMIT database to evaluate the performance of our system. Our results show that the as-proposed system is robust for gender classification across a wide range of noise levels.

关键词： Neurons Encoding Feature extraction Robustness Biological neural networks Task analysis Spectrogram

来源：评论

学校读者我要写书评

暂无评论

Deep technology Tracing for High-Tech Companies

Deep Technology Tracing for High-Tech Companies

引用

IEEE International Conference on Data Mining (ICDM)

作者： Han Wu Kun Zhang Guangyi Lv Qi Liu Runlong Yu Weihao Zhao Enhong Chen Jianhui Ma Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China

Technological change and innovation are vitally important, especially for high-tech companies. However, factors influencing their future research and development (R&D) trends are both complicated and various, leading it a quite difficult task to make technology tracing for high-tech companies. To this end, in this paper, we develop a novel data-driven solution, i.e., Deep technology Forecasting (DTF) framework, to automatically find the most possible technology directions customized to each high-tech company. Specially, DTF consists of three components: Potential Competitor Recognition (PCR), Collaborative technology Recognition (CTR), and Deep technology Tracing (DTT) neural network. For one thing, PCR and CTR aim to capture competitive relations among enterprises and collaborative relations among technologies, respectively. For another, DTT is designed for modeling dynamic interactions between companies and technologies with the above relations involved. Finally, we evaluate our DTF framework on real-world patent data, and the experimental results clearly prove that DTF can precisely help to prospect future technology emphasis of companies by exploiting hybrid factors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Crowdfunding dynamics tracking: A reinforcement learning approach

arXiv

引用

arXiv 2019年

作者： Wang, Jun Zhang, Hefu Liu, Qi Pan, Zhen Tao, Hanqing Anhui Province Key Laboratory of Big Data Analysis and Application School of Computer Science and Technology University of Science and Technology of China

Recent years have witnessed the increasing interests in research of crowdfunding mechanism. In this area, dynamics tracking is a significant issue but is still under exploration. Existing studies either fit the fluctuations of time-series or employ regularization terms to constrain learned tendencies. However, few of them take into account the inherent decision-making process between investors and crowdfunding dynamics. To address the problem, in this paper, we propose a Trajectory-based Continuous Control for Crowdfunding (TC3) algorithm to predict the funding progress in crowdfunding. Specifically, actor-critic frameworks are employed to model the relationship between investors and campaigns, where all of the investors are viewed as an agent that could interact with the environment derived from the real dynamics of campaigns. Then, to further explore the in-depth implications of patterns (i.e., typical characters) in funding series, we propose to subdivide them into fast-growing and slow-growing ones. Moreover, for the purpose of switching from different kinds of patterns, the actor component of TC3 is extended with a structure of options, which comes to the TC3-Options. Finally, extensive experiments on the Indiegogo dataset not only demonstrate the effectiveness of our methods, but also validate our assumption that the entire pattern learned by TC3-Options is indeed the U-shaped one. Copyright © 2019, The Authors. All rights reserved.

关键词： Crowdsourcing

来源：评论

学校读者我要写书评

暂无评论

Pasadena: Perceptually Aware and Stealthy Adversarial Denoise Attack

arXiv

引用

arXiv 2020年

作者： Cheng, Yupeng Guo, Qing Juefei-Xu, Felix Feng, Wei Lin, Shang-Wei Lin, Weisi Liu, Yang Nanyang Technological University Singapore Alibaba Group United States School of Computer Science and Technology College of Intelligence and Computing Tianjin Key Laboratory of Cognitive Computing and Application Tianjin University Tianjin300305 China State Administration of Cultural Heritage China

Image denoising can remove natural noise that widely exists in images captured by multimedia devices due to low-quality imaging sensors, unstable image transmission processes, or low light conditions. Recent works also find that image denoising benefits the high-level vision tasks, e.g., image classification. In this work, we try to challenge this common sense and explore a totally new problem, i.e., whether the image denoising can be given the capability of fooling the state-of-the-art deep neural networks (DNNs) while enhancing the image quality. To this end, we initiate the very first attempt to study this problem from the perspective of adversarial attack and propose the adversarial denoise attack. More specifically, our main contributions are three-fold: First, we identify a new task that stealthily embeds attacks inside the image denoising module widely deployed in multimedia devices as an image post-processing operation to simultaneously enhance the visual image quality and fool DNNs. Second, we formulate this new task as a kernel prediction problem for image filtering and propose the adversarial-denoising kernel prediction that can produce adversarial-noiseless kernels for effective denoising and adversarial attacking simultaneously. Third, we implement an adaptive perceptual region localization to identify semantic-related vulnerability regions with which the attack can be more effective while not doing too much harm to the denoising. We name the proposed method as Pasadena (Perceptually Aware and Stealthy Adversarial DENoise Attack) and validate our method on the NeurIPS’17 adversarial competition dataset, CVPR2021-AIC-VI: unrestricted adversarial attacks on ImageNet, and Tiny-ImageNet-C dataset. The comprehensive evaluation and analysis demonstrate that our method not only realizes denoising but also achieves a significantly higher success rate and transferability over state-of-the-art attacks. Copyright © 2020, The Authors. All rights reserved.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Weakly supervised histopathology tissue semantic segmentation with multi-scale voting and online noise suppression

引用

Engineering applications of Artificial Intelligence 2025年 156卷

作者： Pan, Xipeng Zhang, Hualong Deng, Huahu Wang, Huadeng Li, Lingqiao Liu, Zhenbing Wang, Lin An, Yajun Lu, Cheng Liu, Zaiyi Han, Chu Lan, Rushi School of Computer Science and Information Security Guilin University of Electronic Technology Guangxi Guilin541004 China Southern Medical University Guangdong Guangzhou510080 China Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangdong Provincial People's Hospital Guangdong Guangzhou510080 China Department of Nasopharyngeal Carcinoma Sun Yat-sen University Cancer Center Guangzhou510060 China International Joint Research Laboratory of Spatio-temporal Information and Intelligent Location Services Guilin University of Electronic Technology Guangxi Guilin541004 China

The development of an Artificial Intelligence (AI) assisted tissue segmentation method of digital pathology images is critical for cancer diagnosis and prognosis. Excellent performance has been achieved with the current fully supervised segmentation approach, which relies on a huge number of annotated data. However, drawing dense pixel-level annotations on the giga-pixel whole slide image (WSI) is extremely time-consuming and labor-intensive. To this end, we propose a tissue segmentation method using only patch-level classification labels to reduce such annotation burden and significantly improve the quality of the pseudo-masks. We introduce a framework with two phases of classification and segmentation. In the classification phase, we propose a multi-scale voting method on the Class Activation Map (CAM) based model to obtain more stable pseudo masks. In the segmentation phase, an Online Noise Suppression Strategy (ONSS) is proposed to encourage the model to focus on more reliable signals in the pseudo mask rather than noisy signals. Extensive experiments on two weakly supervised pathology image tissue segmentation datasets Lung Adenocarcinoma (LUAD-HistoSeg) and Breast Cancer Semantic Segmentation (BCSS-WSSS) demonstrate our model outperforms state-of-the-art weakly-supervised semantic segmentation (WSSS) methods using patch-level labels. Furthermore, our method exhibits superior generalization ability compared to other models, and demonstrates promising adaptation performance on unseen domains with only small amounts of data. © 2025 Elsevier Ltd

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Superior performance of ZnGaO solar-blind photodetectors by Implementing TFT structure and tunable ZnO cycle ratio

引用

Materials Today Chemistry 2024年 38卷

作者： Fan, Hui-Chen Wang, Chen Xu, Yi-Hong Fan, Teng-Min Huang, Pao-Hsun Ruan, Yu-Jiao Wu, Ting-Zhu Kuo, Hao-Chung Wuu, Dong-Sing Lai, Feng-Min Lien, Shui-Yang Xiamen Key Laboratory of Development and Application for Advanced Semiconductor Coating Technology School of Opto-Electronic and Communication Engineering Xiamen University of Technology Xiamen 361024 China School of Ocean Information Engineering Jimei University Jimei District Xiamen 361021 Fujian China National Measurement and Testing Center for Flat Panel Display Industry Xiamen Institute of Measurement and Testing Xiamen 361024 China National Integrated Circuit Industry and Education Integration Innovation Platform Department of Electronic Science Xiamen University Xiamen 361005 China Department of Photonics & Graduate Institute of Electro-Optical Engineering College of Electrical and Computer Engineering National Yang Ming Chiao Tung University Hsinchu 30010 Taiwan Department of Applied Materials and Optoelectronic Engineering National Chi Nan University Nantou 54561 Taiwan Department of Biomedical Engineering Da-Yeh University Changhua 51591 Taiwan

In this study, plasma enhanced atomic layer deposition was used to prepare ZnGaO thin films with different ZnO cycle ratios from 0 % to 50 %, which were employed to fabricate the MSM-type and TFT-type solar-blind photodetector. The properties of the films and the performance of the solar blind photodetector are greatly affected by the ZnO cycle ratio. Appropriate Zn doping content decreases the ratio of oxygen vacancies and Zn interstitials in Ga₂O₃, while excessive Zn doping degrades the quality of the film. Compared to MSM-type photodetector, an obviously improved performance of ZnGaO TFT-type photodetector has been observed due to its self-amplified photocurrent and the reduced dark current controlled by gate voltage. At a ZnO cycle ratio of 40 %, a superior performance of TFT-type solar-blind photodetector with an ultra-low dark current ∼4.5 × 10⁻¹³ A, an excellent light switching ratio ∼2.2 × 10⁷, and an outstandingly high sensitivity ∼34.31 A/W has been obtained. These results demonstrate that the incorporation of Zn dopants can effectively modulate the performance of Ga₂O₃ TFT-type photodetector providing a great perspective for cost-effective and high-performance solar-blind PDs. © 2024 Elsevier Ltd

关键词： High performance Plasma enhanced atomic layer deposition TFT-Type solar-blind photodetector ZnGaO film ZnO cycle ratio

来源：评论

学校读者我要写书评

暂无评论

Synaptic learning with augmented spikes

arXiv

引用

arXiv 2020年

作者： Yu, Qiang Song, Shiming Ma, Chenxiang Pan, Linqiang Tan, Kay Chen Tianjin Key Laboratory of Cognitive Computing and Application College of Intelligence and Computing Tianjin University Tianjin China Key Laboratory of Image Information Processing and Intelligent Control Institute of Artificial Intelligence School of Artificial Intelligence and Automation Huazhong University of Science and Technology Wuhan China The Department of Computer Science City University of Hong Kong Hong Kong

Traditional neuron models use analog values for information representation and computation, while all-or-nothing spikes are employed in the spiking ones. With a more brain-like processing paradigm, spiking neurons are more promising for improvements on efficiency and computational capability. They extend the computation of traditional neurons with an additional dimension of time carried by all-or-nothing spikes. Could one benefit from both the accuracy of analog values and the time-processing capability of spikes? In this paper, we introduce a concept of augmented spikes to carry complementary information with spike coefficients in addition to spike latencies. New augmented spiking neuron model and synaptic learning rules are proposed to process and learn patterns of augmented spikes. We provide systematic insight into the properties and characteristics of our methods, including classification of augmented spike patterns, learning capacity, construction of causality, feature detection, robustness and applicability to practical tasks such as acoustic and visual pattern recognition. The remarkable results highlight the effectiveness and potential merits of our methods. Importantly, our augmented approaches are versatile and can be easily generalized to other spike-based systems, contributing to a potential development for them including neuromorphic computing. Copyright © 2020, The Authors. All rights reserved.

关键词： Neurons

来源：评论

学校读者我要写书评

暂无评论

An improved method of algal-bloom discrimination in Taihu Lake using Sentinel-1A data 6

An improved method of algal-bloom discrimination in Taihu La...

引用

6th Asia-Pacific Conference on Synthetic Aperture Radar, APSAR 2019

作者： Wu, Lin Sun, Mengwei Min, Lin Zhao, Jianhui Li, Ning Guo, Zhengwei College of Computer and Information Engineering Henan University Kaifeng475004 China Henan Key Laboratory of Big Data Analysis and Processing Henan University Kaifeng475004 China Henan Engineering Research Center of Intelligent Technology and Application Kaifeng475004 China

ISBN: (纸本)9781728129129

The algal bloom is a prominent manifestation of water pollution. Synthetic aperture radar (SAR) shows an advantage in water monitoring due to its characteristic of all-time and all-weather. The water regions where algae gather present dark in SAR image. However, dark regions may also be caused by other factors, such as low wind. This paper proposes an improved algal bloom discrimination method based on Artificial Neural Network (ANN) to recognize the dark regions of algal bloom. Taihu Lake is chosen as the research area in this study because of its serious bloom in recent years. By means of quasi-synchronous optical images, the dark region database of SAR images labeled algal bloom and non-algal bloom are obtained. Then the segmentation algorithm and region growing algorithm are used to acquire the feature from dark regions, and divided into training feature set and testing feature set. Finally, the training and testing feature set are used for ANN-based discrimination model construction and verification. According the experimental results, the overall accuracy reaches 80%, which indicates that ANN model has a good applicability in algal bloom recognition of SAR image. © 2019 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Transient Fault Diagnosis Algorithm based on Feature Classification and Its application in High-Power Laser facility

Transient Fault Diagnosis Algorithm based on Feature Classif...

引用

Chinese Control and Decision Conference, CCDC

作者： Qiang Ren Hang Gao Hailei Ren Baoran An Tianyou Yun Xiaoli Wang Institute of Computer Application China Academy of Engineering Physics Mianyang State Key Laboratory of Complex Electromagnetic Environmental Effects on Electronics and Information System College of Electronic Science and Technology National University of Defense Technology Changsha Institute of Fluid Physics China Academy of Engineering Physis MianYang China

ISBN: (数字)9781728158556

ISBN: (纸本)9781728158563

High-power laser facility plays an import role both in scientific research and military application, and the occurred faults are usually transient. In this paper, a transient fault diagnosis algorithm based on feature classification is proposed, and the algorithm has been applied in the high-power laser facility. The method extracts the features of input data with an autoencoder network. And based on the extracted features, a multi-layer full connection neural network is adopted for fault diagnosis. By comparing the results with those obtained by traditional fault diagnosis methods, the performance of the algorithm has been verified.

关键词： Feature extraction Fault diagnosis Circuit faults Neural networks Principal component analysis Kernel Transient analysis

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：