检索结果-内蒙古大学图书馆

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Heming Wang DeLiang Wang Department of Computer Science and Engineering The Ohio State University USA Center for Cognitive and Brain Sciences The Ohio State University USA

Deep learning based speech enhancement has achieved remarkable success, but challenges remain in low signal-to-noise ratio (SNR) nonstationary noise scenarios. In this study, we propose to incorporate diffusion-based learning into an enhancement model and improve robustness in extremely noisy conditions. Specifically, a frequency-domain diffusion-based generative module is employed, and it accepts the enhanced signal obtained from a time-domain supervised enhancement module as an auxiliary input to learn to recover clean speech spectrograms. Experimental results on the TIMIT dataset demonstrate the advantage of this approach and show better enhancement performance over other strong baselines in both -5 and -10 dB SNR noisy conditions.

关键词： Training Deep learning Frequency-domain analysis Speech enhancement Robustness Noise measurement Background noise

来源：评论

学校读者我要写书评

暂无评论

A Multi-UAV Charging Station Enabling Free Landing by Grid Pattern Transmitter

A Multi-UAV Charging Station Enabling Free Landing by Grid P...

引用

Annual IEEE Conference on Applied Power Electronics Conference and Exposition (APEC)

作者： Jungho Kim Hyunkyeong Jo Seoktae Seo Bonyoung Lee Hyungki Min Franklin Bien School of Electrical and Computer Engineering Ulsan National Institute of Science and Technology Ulsan Korea Program in Information & Communication Technology Convergence Ulsan National Institute of Science and Technology Ulsan Korea

ISBN: (数字)9798331516116

ISBN: (纸本)9798331516123

This paper proposed a novel landing platform with wireless power transfer (WPT) to charge multiple unmanned aerial vehicles (UAVs) without restrictions on landing locations or alignment. A transmitter is designed using wires to charge a wide area. To overcome the disadvantages of system made by wires, a grid pattern transmitter and double power line are proposed. These structures provide multiple charging with free positioning of receivers and excite the transmitter lines in wide area properly. The optimized system is composed of 0.15m square cells, providing a 0.9 m by 0.9 m charging area. By experiment, it can efficiently charge four 0.1 m square receivers located anywhere in transmitter. The overall efficiency of four receivers is 37.5%. Also, the experimental results verify that it is unnecessary to align the receiver within the cell, as it has a 2.62% error due to rotation and a 2.94 % error due to position within the cell.

关键词： Coils Magnetic flux leakage Transmitters Wires Windings Receivers Wireless power transfer Autonomous aerial vehicles Current distribution Power electronics

来源：评论

学校读者我要写书评

暂无评论

Pain Classification Using Statistical Feature Extraction Using Machine Learning Approach: A Pilot Study

Pain Classification Using Statistical Feature Extraction Usi...

引用

IEEE International Conference on Control System, Computing and engineering (ICCSCE)

作者： Sudimanto Agung Trisetyarso Iman Herwidiana Kartowisastro Widodo Budiharto Computer Science Department BINUS Graduate Program Computer Science Bina Nusantara University Jakarta Indonesia Mathematics Department School of Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department BINUS Graduate Program Computer Science Computer Engineering Department Bina Nusantara University Jakarta Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jl. K. H. Syahdan No. 9 Kemanggisan Palmerah Jakarta Indonesia

Electrodermal activity (EDA) is a general term for all electrical phenomena occurring on the skin, both passive and active. EDA measurements are used by researchers to measure levels of stress, emotion, mental strain, and so on. Measuring human stress levels, emotions, and mental strain are generally associated with the skin conductance response. The function GSR sensor is not only used to read people’s psychology but also can be used as a pain sensor used to read the degree of pain in the skin. This pilot study uses sample data from ***. The *** data is galvanic skin response sensor data. The output of this sensor is the conductivity value that occurs in the skin. The data obtained from *** will be extracted using the mean, standard deviation, maximum, minimum, RMS, skewness, and peak-to-peak characteristics. The extracted functions are selected using the forward selection method. The results of the feature selection are three features with an accuracy percentage greater than 50%, namely the mean feature, the RMS feature, and the skewness feature. The machine learning models used are bagged tree, SVM, and K-NN models. Of the three models used, the bagged tree model has the highest accuracy rate, at 98.05%, with an F1 score is 0.9807. The KNN model with k=10 has the lowest level of accuracy compared to other models, at 96.75%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Hierarchical B-frame Video Coding Using Two-Layer CANF without Motion Coding

arXiv

引用

arXiv 2023年

作者： Alexandre, David Hang, Hsueh-Ming Peng, Wen-Hsiao Electrical Engineering and Computer Science International Graduate Program Institute of Electronics Department of Computer Science National Yang Ming Chiao Tung University Hsinchu Taiwan

Typical video compression systems consist of two main modules: motion coding and residual coding. This general architecture is adopted by classical coding schemes (such as international standards H.265 and H.266) and deep learning-based coding schemes. We propose a novel B-frame coding architecture based on two-layer Conditional Augmented Normalization Flows (CANF). It has the striking feature of not transmitting any motion information. Our proposed idea of video compression without motion coding offers a new direction for learned video coding. Our base layer is a low-resolution image compressor that replaces the full-resolution motion compressor. The low-resolution coded image is merged with the warped high-resolution images to generate a high-quality image as a conditioning signal for the enhancement-layer image coding in full resolution. One advantage of this architecture is significantly reduced computational complexity due to eliminating the motion information compressor. In addition, we adopt a skip-mode coding technique to reduce the transmitted latent samples. The rate-distortion performance of our scheme is slightly lower than that of the state-of-the-art learned B-frame coding scheme, B-CANF, but outperforms other learned B-frame coding schemes. However, compared to B-CANF, our scheme saves 45% of multiply–accumulate operations (MACs) for encoding and 27% of MACs for decoding. The code is available at https://***. Copyright © 2023, The Authors. All rights reserved.

关键词： Image coding

来源：评论

学校读者我要写书评

暂无评论

Enhancing Security Information and Event Management (SIEM) by Incorporating Machine Learning for Cyber Attack Detection

Enhancing Security Information and Event Management (SIEM) b...

引用

International Conference on Computing, engineering, and Design (ICCED)

作者： Media Anugerah Ayu Dimas Erlangga Teddy Mantoro Dini Handayani Department of Computer Science Faculty of Engineering and Technology Sampoerna University Jakarta Indonesia Postgraduate Program - School of Computer Science Nusa Putra University - Sukabumi Indonesia

Network security is a crucial component of Information Technology, yet organizations continue to grapple with meeting established security benchmarks. Given the rise in cyber-attacks and the continuous emergence of new attack types, it’s practically infeasible to persistently update attack patterns or signatures within security parameters. Key tools such as Intrusion Detection Systems (IDS) and Security Information and Event Management (SIEM) are instrumental in monitoring network traffic and identifying potential threats. However, these tools face limitations, such as the high volume of alerts produced by IDS and the use of rule-based method, also the inability of SIEM tools to analyze logs comprehensively to identify inappropriate activities. This research has conducted anomaly detection using machine learning process to classify cyber-attacks network flow collected from IDS that installed incident network infrastructure. The analysis of IDS using machine learning, integrated with SIEM. The algorithm used in this research was Random Forest Classifier using CSE-CID-IDS2018 dataset pre-processed with Principal Component Analysis (PCA). Results of the experiments show that Random Forest Classifier Model, when combined with Principal Component Analysis (PCA), yields the most commendable results when applied to a 70/30 training/testing data ratio with accuracy of 0.99953.

关键词：

来源：评论

学校读者我要写书评

暂无评论

In-Solution Germanium Selenide Nanosheets as Charge Trapping Layer in Flash Memories

In-Solution Germanium Selenide Nanosheets as Charge Trapping...

引用

IEEE Silicon Nanoelectronics Workshop (SNW)

作者： Nazek El-Atab Bashayr Alqahtani Program of Electrical and Computer Engineering Computer Electrical Mathematical Sciences and Engineering King Abdullah University of Science and Technology Thuwal Saudi Arabia

Germanium selenide (GeSe) is a highly promising material with several attractive characteristics, particularly in the field of ferroelectric and phase-change memories due to its outstanding electronic behavior. However, the potential of GeSe as a charge-trapping layer in flash memory has received less attention. Herein, the fabrication of a nonvolatile MOS memory device using GeSe nanosheets as a charge-trapping layer was demonstrated and the materials flakes were examined extensively. The electrical performance of the memory device was investigated. Intriguingly, it exhibited an extraordinarily wide memory window of 9 V under ±10 V electrical biasing. Additionally, the devices presented high endurance of $10^{4}$ programming and erasing cycles, and reliable charge storage of only 56% loss after 10 years.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-granularity Textual Adversarial Attack with Behavior Cloning

Multi-granularity Textual Adversarial Attack with Behavior C...

引用

2021 Conference on Empirical Methods in Natural Language Processing, EMNLP 2021

作者： Chen, Yangyi Su, Jin Wei, Wei Cognitive Computing and Intelligent Information Processing Laboratory School of Computer Science and Technology Huazhong University of Science and Tehchnology China School of Software Engineering Huazhong University of Science and Tehchnology China

ISBN: (纸本)9781955917094

Recently, the textual adversarial attack models become increasingly popular due to their successful in estimating the robustness of NLP models. However, existing works have obvious deficiencies. (1) They usually consider only a single granularity of modification strategies (e.g. word-level or sentence-level), which is insufficient to explore the holistic textual space for generation;(2) They need to query victim models hundreds of times to make a successful attack, which is highly inefficient in practice. To address such problems, in this paper we propose MAYA, a MultigrAnularitY Attack model to effectively generate high-quality adversarial samples with fewer queries to victim models. Furthermore, we propose a reinforcement-learning based method to train a multi-granularity attack agent through behavior cloning with the expert knowledge from our MAYA algorithm to further reduce the query times. Additionally, we also adapt the agent to attack black-box models that only output labels without confidence scores. We conduct comprehensive experiments to evaluate our attack models by attacking BiLSTM, BERT and RoBERTa in two different black-box attack settings and three benchmark datasets. Experimental results show that our models achieve overall better attacking performance and produce more fluent and grammatical adversarial samples compared to baseline models. Besides, our adversarial attack agent significantly reduces the query times in both attack settings. Our codes are released at https://***/Yangyi-Chen/MAYA. © 2021 Association for Computational Linguistics

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Systematic Biases in LLM Simulations of Debates

Systematic Biases in LLM Simulations of Debates

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Taubenfeld, Amir Dover, Yaniv Reichart, Roi Goldstein, Ariel The Hebrew University of Jerusalem School of Computer Science and Engineering Israel Google Research United States The Hebrew University Business School Jerusalem Israel Federmann Center for the Study of Rationality Hebrew University Jerusalem Israel Faculty of Data and Decision Sciences Technion Israel Department of Cognitive and Brain Sciences Hebrew University Jerusalem Israel

ISBN: (纸本)9798891761643

The emergence of Large Language Models (LLMs), has opened exciting possibilities for constructing computational simulations designed to replicate human behavior accurately. Current research suggests that LLM-based agents become increasingly human-like in their performance, sparking interest in using these AI agents as substitutes for human participants in behavioral studies. However, LLMs are complex statistical learners without straightforward deductive rules, making them prone to unexpected behaviors. Hence, it is crucial to study and pinpoint the key behavioral distinctions between humans and LLM-based agents. In this study, we highlight the limitations of LLMs in simulating human interactions, particularly focusing on LLMs' ability to simulate political debates on topics that are important aspects of people's day-to-day lives and decision-making processes. Our findings indicate a tendency for LLM agents to conform to the model's inherent social biases despite being directed to debate from certain political perspectives. This tendency results in behavioral patterns that seem to deviate from well-established social dynamics among humans. We reinforce these observations using an automatic self-fine-tuning method, which enables us to manipulate the biases within the LLM and demonstrate that agents subsequently align with the altered biases. These results underscore the need for further research to develop methods that help agents overcome these biases, a critical step toward creating more realistic simulations. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Feasibility Study of Semi-supervised Intrusion Detection Systems for Industrial Control Systems

Feasibility Study of Semi-supervised Intrusion Detection Sys...

引用

International Conference on Applied System Innovation (ICASI)

作者： Fu-Nie Loo Tung-Shou Chen Chia-Wei Tsai Jason Lin Chun-Wei Yang Department of Computer Science and Information Engineering National Taichung University of Science and Technology Taichung City Taiwan Department of Computer Science and Engineering National Chung Hsing University Taichung City Taiwan Master Program for Digital Health Innovation China Medical University Taichung City Taiwan

ISBN: (数字)9798350394924

ISBN: (纸本)9798350394931

Industrial Control systems (ICS) automate industrial processes but also introduces cybersecurity threats. Intrusion Detection System (IDS) are crucial for detecting cyber-attacks on ICS, yet zero-day attacks are often inefficiently detecting with supervised learning. This study employs semi-supervised learning using one-class SVM, isolation forest, and Local Outlier Factor (LOF), to train IDS models. Utilizing dataset collected from a self-build virtual ICS environment, the study demonstrates the feasibility of these models in detecting common attack like Injection, ARP, and Man-in-the-Middle.

关键词： Training Support vector machines Technological innovation Supervised learning Intrusion detection Forestry Semisupervised learning

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis With N-Gram Preprocessing for Online-Shopping Reviews in Indonesian Language

Sentiment Analysis With N-Gram Preprocessing for Online-Shop...

引用

International Conference on Computing, engineering, and Design (ICCED)

作者： Jelita Asian Oktavianus Tegar N.K Putra Media Anugerah Ayu Teddy Mantoro Postgraduate Program - School of Computer Science Nusa Putra University Sukabumi Indonesia Department of Computer Science Faculty of Engineering and Technology Sampoerna University Jakarta Indonesia

Currently, online Shopping platforms have grown significantly, especially during the COVID-19 pandemic. This condition motivates the need for analyzing how the users/customers’ opinions on using such platform. Sentiment analysis, as a process of detecting, extracting, and classifying users’ opinions and attitudes toward specific topics, is a good tool for the required analysis. This study aims to evaluate the performance of machine learning approach which combined with N-Gram technique in doing sentiment analysis. The dataset used in this study comes from scraping reviews in Bahasa Indonesia regarding the Shopee Apps. In this study, $\mathrm{N}=2$ for the N-Gram was employed in the preprocessing process. Our main goal is to investigate whether the performance of machine learning in doing sentiment analysis can be improved by adding the N-Gram technique in its preprocessing. This work applied the Naive Bayes Classifier and k-Nearest Neighbor with $K=11$ as the machine learning algorithms. The best accuracy in this study was achieved by Naive Bayes Classifier after applying N-Gram Terms $(N=2)$ with Split Validation (8:2), which is $\mathbf{97.26\%}$.

关键词：

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：