检索结果-内蒙古大学图书馆

Siberian Symposium on data Science and Engineering (SSDSE)

作者： Khazankin, Grigory R. Komarov, Sergey Kovalev, Danila Barsegyan, Artur Likhachev, Alexander Novosibirsk State Univ Stream Data Analyt & Machine Learning Lab Novosibirsk Russia

ISBN: (纸本)9781538615935

To solve the problems associated with large data volume real-time processing, heterogeneous systems using various computing devices are increasingly used. The characteristic of solving this class of problems is related to the fact that there are two directions for improving methods of real-time data analysis: the first is the development of algorithms and approaches to analysis, and the second is the development of hardware and software. This article reviews the main approaches to the architecture of a hardware-software solution for traffic capture and deep packet inspection (DPI) in data transmission networks with a bandwidth of 80 Gbit/s and higher. At the moment there are software and hardware tools that allow designing the architecture of capture system and deep packet inspection: Using only the central processing unit (CPU);Using only the graphics processing unit (GPU);Using the central processing unit and graphics processing unit simultaneously (CPU + GPU). In this paper, we consider these key approaches. Also attention is paid to both hardware and software requirements for the architecture of solutions. Pain points and remedies are described.

关键词： Napatech Mellanox DPDK PeerDirect DPI Nvidia CUDA GPUDirect Zero-copy DMA

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence for Imaging Diagnostics in Neurosurgery

Artificial Intelligence for Imaging Diagnostics in Neurosurg...

引用

IEEE Region 8 International Conference on Computational Technologies in Electrical and Electronics Engineering, SIBIRCON

作者： Andrey Yu. Letyagin Sergey K. Golushko Jamil A. Rzaev Mihail E. Amelin Evgeniy N. Pavlovsky Bair N. Tuchinov Evgeniya V. Amelina Galina I. Moisak Ekaterina G. Bulgakova Liana O. Degtyareva Novosibirsk State University Novosibirsk Russia Federal State Budget Institution Federal Center of Neurosurgery Novosibirsk Russia FSBI “Federal Neurosurgical Center” Novosibirsk Russia Stread Data Analytics & Machine Learning lab. Novosibirsk State University Novosibirsk Russia

Improving the accuracy and timeliness of medical imaging in everyday clinical neurosurgery practice is an urgent problem for the health care of all countries. This problem will be solved with the help of artificial intelligence (AI). By performing the search and contouring of “suspicious” pathological objects, making automatic measurements on 3D medical images (MRI, CT), formalizing quantitative indicators and compiling medical statistical reports, AI can solve diagnostic problems in radiology and medical imaging. The effect of introducing AI will be at least a more accurate and quick interpretation of images, implemented through a set of decision-making tools for medical specialists.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Reducing over-smoothness in speech synthesis using Generative Adversarial Networks

arXiv

引用

arXiv 2018年

作者： Sheng, Leyuan Pavlovskiy, Evgeniy N. Department of Mathematics and Mechanics Novosibirsk State University Novosibirsk Russia Stream Data Analytics and Machine Learning laboratory Novosibirsk State University Novosibirsk Russia

Speech synthesis is widely used in many practical applications. In recent years, speech synthesis technology has developed rapidly. However, one of the reasons why synthetic speech is unnatural is that it often has over-smoothness. In order to improve the naturalness of synthetic speech, we first extract the mel-spectrogram of speech and convert it into a real image, then take the over-smooth mel-spectrogram image as input, and use image-to-image translation Generative Adversarial Networks(GANs) framework to generate a more realistic mel-spectrogram. Finally, the results show that this method greatly reduces the over-smoothness of synthesized speech and is more close to the mel-spectrogram of real speech. Copyright © 2018, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Structure of Attention in a Transformer Language Model

arXiv

引用

arXiv 2019年

作者： Vig, Jesse Belinkov, Yonatan Palo Alto Research Center Machine Learning and Data Science Group Interaction and Analytics Lab Palo AltoCA United States Harvard John A. Paulson School of Engineering and Applied Sciences MIT Computer Science and Artificial Intelligence Laboratory CambridgeMA United States

The Transformer is a fully attention-based alternative to recurrent networks that has achieved state-of-the-art results across a range of NLP tasks. In this paper, we analyze the structure of attention in a Transformer language model, the GPT-2 small pretrained model. We visualize attention for individual instances and analyze the interaction between attention and syntax over a large corpus. We find that attention targets different parts of speech at different layer depths within the model, and that attention aligns with dependency relations most strongly in the middle layers. We also find that the deepest layers of the model capture the most distant relationships. Finally, we extract exemplar sentences that reveal highly specific patterns targeted by particular attention heads. Copyright © 2019, The Authors. All rights reserved.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

FRiS-censoring of reference sample in face recognition task by deep neural networks

FRiS-censoring of reference sample in face recognition task ...

引用

Siberian Symposium on data Science and Engineering (SSDSE)

作者： Alyamkin, Sergey A. Nikolenko, Nikita A. Pavlovskiy, Evgeniy N. Dyubanov, Vladimir V. Novosibirsk State Univ Stream Data Analyt & Machine Learning Lab Novosibirsk Russia Expasoft LLC Novosibirsk Russia

To increase face recognition quality in video surveillance system an approach of censoring incoming photos based on FRiS function is presented.

ISBN: (纸本)9781538615935

To increase face recognition quality in video surveillance system an approach of censoring incoming photos based on FRiS function is presented.

关键词： convolution neural network FRiS censoring face recognition

来源：评论

学校读者我要写书评

暂无评论

machine learning Techniques for Stress Prediction in Working Employees

Machine Learning Techniques for Stress Prediction in Working...

引用

IEEE International Conference on Computational Intelligence and Computing Research

作者： U Srinivasulu Reddy Aditya Vivek Thota A Dharun Department of Computer Applications Machine Learning and Data Analytics Lab National Institute of Technology Trichy ECE National Institute of Technology Trichy

ISBN: (纸本)9781538615096

Stress disorders are a common issue among working IT professionals in the industry today. With changing lifestyle and work cultures, there is an increase in the risk of stress among the employees. Though many industries and corporates provide mental health related schemes and try to ease the workplace atmosphere, the issue is far from control. In this paper, we would like to apply machine learning techniques to analyze stress patterns in working adults and to narrow down the factors that strongly determine the stress levels. Towards this, data from the OSMI mental health survey 2017 responses of working professionals within the tech-industry was considered. Various machine learning techniques were applied to train our model after due data cleaning and preprocessing. The accuracy of the above models was obtained and studied comparatively. Boosting had the highest accuracy among the models implemented. By using Decision Trees, prominent features that influence stress were identified as gender, family history and availability of health benefits in the workplace. With these results, industries can now narrow down their approach to reduce stress and create a much comfortable workplace for their employees.

关键词： data handling decision trees DP management employee welfare ergonomics gender issues health care learning (artificial intelligence) occupational stress organisational aspects personnel psychology risk analysis computer installation management Decision Trees Organizational aspects gender issue data processing stress at work Ergonomics Psychology Risk Management risk analysis Personnel Employees Health Care data cleaning learning methods

来源：评论

学校读者我要写书评

暂无评论

Real-time Mental State Recognition using a Wearable EEG

Real-time Mental State Recognition using a Wearable EEG

引用

Annual International Conference of the IEEE Engineering in Medicine and Biology Society

作者： Robert Richer Nan Zhao Judith Amores Bjoern M. Eskofier Joseph A. Paradiso Machine Learning and Data Analytics Lab Friedrich-Alexander-Universitt Erlangen-Nrnberg (FAU) Erlangen Germany Massachusetts Institute of Technology Media Lab Cambridge MA US

The increasing quality and availability of low-cost EEG systems offer new possibilities for non-medical purposes. Existing openly available algorithms to assess the user's mental state in real-time have been mainly performed with medical-grade equipment. In this paper, an approach to assess the user's Focus or Relax states in real-time using a consumer-grade, wearable EEG headband is evaluated. One naive measure and four entropy-based measures, computed using relative frequency band powers in the EEG signal, were introduced. Classifiers for relax and focus state detection, based on the estimation of probability distributions, were developed and evaluated in a user study. Results showed that the Tsallis entropy-based measure performed best for the Focus score, whereas the Renyi measure performed best for the Relax score. Sensitivities of 82.0% and 80.4% with specificities of 82.8% and 80.8% were achieved for the Focus and Relax scores, respectively. The results demonstrated the possibilities of using a wearable EEG system for real-time mental state recognition.

关键词： Electroencephalography Entropy Real-time systems Task analysis Histograms Frequency measurement Sensitivity

来源：评论

学校读者我要写书评

暂无评论

Supervised learning for Yaw Orientation Estimation

Supervised Learning for Yaw Orientation Estimation

引用

International Conference on Indoor Positioning and Indoor Navigation (IPIN)

作者： Tobias Feigl Christopher Mutschler Michael Philippsen Machine Learning and Information Fusion Group Fraunhofer Institute for Integrated Circuits IIS Nuremberg Germany Machine Learning and Data Analytics Lab Friedrich-Alexander University Erlangen-Nürnberg Erlangen Germany Programming Systems Group

With free movement and multi-user capabilities, there is demand to open up Virtual Reality (VR) for large spaces. However, the cost of accurate camera-based tracking grows with the size of the space and the number of users. No-pose (NP) tracking is cheaper, but so far it cannot accurately and stably estimate the yaw orientation of the user's head in the long-run. Our novel yaw orientation estimation combines a single inertial sensor located at the human's head with inaccurate positional tracking. We exploit that humans tend to walk in their viewing direction and that they also tolerate some orientation drift. We classify head and body motion and estimate heading drift to enable low-cost long-time stable head orientation in NP tracking on 100 m×100 m. Our evaluation shows that we estimate heading reasonably well.

关键词： Magnetic heads Feature extraction Acceleration Tracking Head Accelerometers Estimation

来源：评论

学校读者我要写书评

暂无评论

A Mobile Solution for Rhythmic Auditory Stimulation Gait Training

A Mobile Solution for Rhythmic Auditory Stimulation Gait Tra...

引用

Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)

作者： Katharina Aholt Christine F. Martindale Arne Küderle Heiko Gaßner Till Gladow Javier Rojo Samanta Villanueva-Mascato Jochen Klucken Maria Teresa Arredondo Waldmeyer Bjoern M. Eskofier Machine Learning and Data Analytics Lab Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU) Erlangen Germany University Hospital Erlangen Germany Escuela Técnica Superior de Ingenieros de Telecomunicación Universidad Politécnica de Madrid Madrid Spain Life Supporting Technologies Universidad Politécnica de Madrid Spain

Recent studies showed that Parkinson's disease (PD) patients improved their gait parameters while walking with rhythmic auditory stimulation (RAS). They achieved a longer stride length, a reduced stride time variability and a higher walking speed. Combining RAS with mobile gait analysis would allow continuous monitoring of RAS effects and gait in natural environments. This paper proposes a mobile solution for home-based assessment of RAS by combining RAS gait training and a mobile system for data acquisition. Existing datasets were used to investigate the cadence of PD patients and to propose suitable frequencies for RAS gait training. The cadence calculation was implemented using a peak detection algorithm, which uses the time difference between two mid-swing events as stride time values. We validated our system as a whole using a cohort of 13 PD patients who performed RAS gait training. The algorithms were also validated against the eGaIT system, a state-of-the-art system, and achieved a mean F 1 score for detected strides of 97.57 % ± 0.86 % and a mean absolute error for the cadence of 0.16 spm ± 0.09 spm. This study lays the ground work for further clinical studies investigating the effectiveness of mobile RAS within a home environment.

关键词： Training Sensors Mobile handsets Parkinson's disease Legged locomotion Gyroscopes data analysis

来源：评论

学校读者我要写书评

暂无评论

Indoor Positioning Using OFDM-Based Visible Light Communication System

Indoor Positioning Using OFDM-Based Visible Light Communicat...

引用

International Conference on Indoor Positioning and Indoor Navigation (IPIN)

作者： Birendra Ghimire Jochen Seitz Christopher Mutschler Precise Positioning and Analytics Department Fraunhofer Institute for Integrated Circuits IIS Nürnberg Germany Machine Learning and Data Analytics Lab Friedrich-Alexander University Erlangen-Nürnberg (FAU) Erlangen Germany

Precise indoor positioning is essential to support emerging applications of location aware mobile computing. Current positioning techniques that use signals transmitted in the Gigahertz region of radio-frequency spectrum do not provide highly accurate position estimates due to multipath propagation of signals. This paper proposes a novel positioning system which uses the entities of visible light communication (VLC) system as anchors and tags. Our technique scales with the number of tags and the VLC network provides both positioning and communication capabilities. The anchor points transmit the OFDM-based VLC signals synchronously, and we estimate the time differences of arrival between anchor points and tags using positioning reference signals embedded into the air-interface of the VLC system. Simulation results show that positioning accuracy of 10 cm or better is possible for over 95% of users if the sampling clock offset is better that 10 ppm, clock jitter is below 1 ps, and a bit resolution of at least 16 bits is available.

关键词： Light emitting diodes OFDM Optical receivers Adaptive optics Optical filters Measurement uncertainty

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：