检索结果-内蒙古大学图书馆

Synergizing spintronics and quaternary logic: a hardware accelerator for neural networks with optimized quantization algorithm

引用

JOURNAL OF SUPERCOMPUTING 2025年第5期81卷 1-28页

作者： Bahmanabadi, Motahareh Amirany, Abdolah Moaiyeri, Mohammad Hossein Jafari, Kian Shahid Beheshti Univ Fac Elect Engn Tehran Iran George Washington Univ Dept Elect & Comp Engn Washington DC USA Univ Sherbrooke Interdisciplinary Inst Technol Innovat 3IT Sherbrooke PQ Canada Univ Sherbrooke Fac Engn 2500 Blvd Univ Sherbrooke PQ Canada

Multi-valued logic (MVL) is a promising solution for high-power consumption and area caused by the limitation of binary systems. Quaternary logic is one of the MVL types more compatible with binary logic. While traditional CMOS technology struggles with achieving multiple threshold voltage levels efficiently, leading to greater complexity and energy inefficiency, CNTFETs offer tunable thresholds ideal for MVL applications. Neural networks can retrieve data from inaccurate data, detect trends, and extract patterns that traditional computing methods or humans find difficult to extract. In artificial neural networks (ANNs), the demand for substantial memory to store numerous weights is a critical consideration. Leveraging emerging technologies like magnetic tunnel junctions (MTJ) for non-volatility and CNTFETs for diverse threshold voltage values presents a groundbreaking stride. This convergence of technologies assures non-volatile storage and facilitates the intricate hardware implementation of MVL systems, marking a pioneering leap forward in crafting the next generation of memory for ANN applications. This paper proposes an algorithm for quantizing neural networks using quaternary logic. According to the proposed algorithm, the circuits required to implement quaternary neural networks are designed. The simulation results demonstrate that the proposed algorithm for quantized neural networks significantly reduces the overall memory requirements compared to their full precision counterparts, with minimal accuracy degradation. Specifically, the accuracy drop is less than 1.67% on CIFAR-10 and 1.22% on CIFAR-100 for ResNet-18, while improvements for MNIST on MLP, CIFAR-10 on LeNet-5, and CIFAR-10 and CIFAR-100 on VGG-16, are even observed.

关键词： Quaternary logic CNTFET MTJ Neural network hardware accelerator quantization algorithm

来源：评论

学校读者我要写书评

暂无评论

Hierarchical quantization algorithm for Deep Learning Network Models

Hierarchical Quantization Algorithm for Deep Learning Networ...

引用

2024 International Conference on Machine Learning and Intelligent Computing, MLIC 2024

作者： Mao, Xiaoqi AnHui Siliepoch Technology Co. Ltd. AnHui HeFei China

Deep learning network models have achieved inspiring performances in various fields such as computer vision, natural language processing, and biomedicine. However, the high computational and storage costs of the models restrain their application in resource-limited situations. However, due to the increased com-plexity and computation of deep neural networks, there are still some challenges in deploying deep learning models into real-world applications for resource-constrained devices. To address this problem, researchers have proposed various quantization algorithms to decrease the expenditure of calculation and storage in deep learning models. This thesis addresses the problem of hierarchical quantiza-tion of deep learning models and proposes a simple hierarchical quantization al-gorithm that aims to effectively reduce the computation and storage requirements of deep learning network models and maintain the accuracy of the models. To demonstrate the effectiveness of the proposed hierarchical quantization method, we conducted experiments on several classical deep learning models. Our exper-iments prove our approach can better maintain the models' accuracy while reduc-ing the storage and computation requirements compared to the traditional quanti-zation algorithms. © 2024 X. Mao.

关键词： computer vision deep learning hierarchical quantization quantization algorithm

来源：评论

学校读者我要写书评

暂无评论

The Impact of quantization Bits on Coherent Detection DAS

引用

JOURNAL OF LIGHTWAVE TECHNOLOGY 2024年第10期42卷 3936-3943页

作者： Ye, Zhenyu Zhang, Shibo Wan, Anchi Zhang, Jialei Wang, Zinan Univ Elect Sci & Technol China Key Lab Opt Fiber Sensing & Commun Chengdu 611731 Peoples R China Ctr Informat Geosci Chengdu 611731 Peoples R China

Distributed acoustic sensing (DAS) is an emerging technology that has gained significance in various domains, including geophysics and security, due to its unique characteristics of long-distance capability, massive sensing points, and large bandwidth. DAS based on coherent detection has been heavily studied in recent years, due to its unique advantages over direct-detection scheme. However, coherent detection DAS usually generates a significant volume of data, since probe signals with wider bandwidth generally have better performance, which presents substantial challenges in data acquisition, storage, transmission and processing, especially for real-time applications. The study of quantization algorithm could be an effective means of addressing these challenges. It should be noted that, due to the impact of quantization noise, signals after quantization exhibit different levels of distortion depending on the quantization bits, resulting in the deterioration of noise floor. This article introduces, to the best of our knowledge, the theory elucidating how quantization bits affect the performance of coherent detection DAS for the first time. It derives a specific formula suitable for quantitative analysis, and the formula's outcomes align with both simulation and experimental results. In addition, this article shows a 50 km long-distance experiment using nonlinear frequency-modulated pulses as the probing signal, further extending the practical application scope of the theory. This article serves as a comprehensive guide for reducing hardware requirements, accelerating data processing to satisfy versatile and demanding application scenarios, and fostering interdisciplinary research.

关键词： quantization (signal) Coherence Phase noise Floors Optical pulses Optical fiber sensors Optical variables control Distributed acoustic sensing coherent detection quantization algorithm noise floor

来源：评论

学校读者我要写书评

暂无评论

Physical-Layer Authentication Using Multiple Channel-Based Features

引用

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 2021年 16卷 2356-2366页

作者： Xie, Ning Chen, Junjie Huang, Lei Shenzhen Univ Coll Elect & Informat Engn Guangdong Key Lab Intelligent Informat Proc Shenzhen 518060 Peoples R China Shenzhen Univ Coll Elect & Informat Engn Shenzhen Key Lab Media Secur Shenzhen 518060 Peoples R China Shenzhen Univ Coll Elect & Informat Engn Guangdong Engn Res Ctr Posit Sensing & Detect Shenzhen 518060 Peoples R China

This paper concerns the problem of authenticating the transmitter without a secret key. In comparison with traditional cryptographic-based authentication mechanisms, the Physical-Layer Authentication (PLA) has the following advantages: high security, low complexity, and high compatibility, since it exploits intrinsic and unique features of the physical layer to authenticate the transmitter rather than using a secret key. The prior channel-based PLA schemes use a quantization algorithm to deal with multiple channel-based features for simplicity. However, there are two main limitations in the prior schemes: performance loss due to quantization error and the difficulty of obtaining the optimal thresholds in closed-form. In this paper, we propose two multiple Channel Impulse Response (CIR) based PLA schemes to effectively overcome the aforementioned limitations of the prior schemes. The first scheme uses multiple CIRs to realize the PLA, which is named as the Multiple CIRs PLA (MCP) scheme. The MCP scheme has better authentication performance than the prior schemes, since it avoids to use a quantization algorithm. The second scheme further improves the authentication performance by exploiting the channel correlation coefficient, which is named as the Enhanced Multiple CIRs PLA (EMCP) scheme. We provide rigorous performance analysis of two proposed schemes. We implemented the proposed schemes and conducted extensive performance comparisons through simulations. Our experimental results show that the closed-form expressions of the theoretical results of the proposed schemes perfectly match the corresponding simulation results. The EMCP scheme has the best authentication performance and the MCP scheme is the second one, whereas the prior scheme is the worst one. As the SNR or the channel correlation coefficient declines, the performance gap among various schemes gradually increases.

关键词： Authentication Programmable logic arrays quantization (signal) Receivers Cryptography MIMO communication Correlation Physical layer authentication channel-based features quantization algorithm performance analysis fading channels

来源：评论

学校读者我要写书评

暂无评论

Multiple Phase Noises Physical-Layer Authentication

引用

IEEE TRANSACTIONS ON COMMUNICATIONS 2022年第9期70卷 6196-6211页

作者： Xie, Ning Xiong, Wei Chen, Junjie Zhang, Peichang Huang, Lei Su, Jian Shenzhen Univ Coll Elect & Informat Engn Guangdong Key Lab Intelligent Informat Proc Shenzhen 518060 Peoples R China Shenzhen Univ Coll Elect & Informat Engn Guangdong Engn Res Ctr Posit Sensing & Detect Shenzhen 518060 Peoples R China Nanjing Univ Informat Sci & Technol Sch Comp Sci Nanjing 210044 Jiangsu Peoples R China

This paper concerns the problem of defending against spoofing attacks without a secret key. We address the problem using the Physical-Layer-Authentication (PLA) because of its high security, low overhead, and high compatibility. However, many PLA schemes have the following limitations: quantization errors, local optimums, and performance loss due to the change in the communication environment. In this paper, two phase-noise-based PLA schemes are proposed to address the limitations of the prior schemes. We denote the first scheme as the Multiple Phase Noises PLA (MPP) scheme, which realizes the PLA by using multiple phase noise innovations. Note that since the MPP scheme avoids using any quantization algorithm, it outperforms the prior schemes on the authentication performance. We denote the second scheme as the Enhanced Multiple Phase Noises PLA (EMPP) scheme, which introduces an artificial random phase to the transmitted symbols at the transmitter to further improve the authentication performance. The theoretical analyses of the proposed schemes over fading channels are provided, where the closed-form expressions are derived. Theoretical comparisons between the proposed schemes and prior schemes are provided. The theoretical analyses and simulation results demonstrated the superiority of the proposed schemes. In comparison with the prior schemes, the MPP scheme achieves 13% authentication-performance gain without demodulation-performance loss, while the EMPP scheme achieves 42% authentication-performance gain with merely 16% demodulation-performance loss.

关键词： Programmable logic arrays Phase noise Authentication quantization (signal) Estimation Demodulation Wireless communication Physical layer authentication phase noise quantization algorithm performance analysis fading channels

来源：评论

学校读者我要写书评

暂无评论

Quantify the weld line and its optimization based on the Taguchi and the moldflow secondary development

Quantify the weld line and its optimization based on the Tag...

引用

2nd International Conference on Advances in Computational Modeling and Simulation (ACMS 2013)

作者： Shan, Zhi Qin, Shuhao Wei, Liangqiang Natl Engn Res Ctr Compounding & Modificat Polymer Guiyang 550014 Peoples R China

ISBN: (纸本)9783037859063

Designed the weld line quantization algorithm based on the moldflow secondary development of technology, obtained the numerical experimental data of weld line with this. Analyzed of the impact of the relationship between the process parameters and weld line, and optimized the process parameters by orthogonal experimental design. These provided an important basis for the forming process design.

关键词： weld line quantization algorithm Taguchi secondary development

来源：评论

学校读者我要写书评

暂无评论

MIRA-ChatGLM: A Fine-Tuned Large Language Model for Intelligent Risk Assessment in Coal Mining

引用

APPLIED SCIENCES-BASEL 2024年第24期14卷 12072页

作者： Sun, Yi Zhang, Chao Wang, Chen Han, Ying Xian Univ Sci & Technol Sch Commun & Informat Engn Xian 710054 Peoples R China

Intelligent mining risk assessment (MIRA) is a vital approach for enhancing safety and operational efficiency in mining. In this study, we introduce MIRA-ChatGLM, which leverages pre-trained large language models (LLMs) for the domain of gas risk assessment in coal mines. We meticulously constructed a dataset specifically designed for mining risk analysis and performed parameter-efficient fine-tuning on the locally deployed GLM-4-9B-chat base model to develop MIRA-ChatGLM. By utilizing consumer-grade GPUs and employing LoRA and various levels of quantization algorithms such as QLoRA, we investigated the impact of different data scales and instruction settings on model performance. The evaluation results show that MIRA-ChatGLM achieved excellent performance with BLEU-4, ROUGE-1, ROUGE-2, and ROUGE-L scores of 84.47, 90.63, 86.88, and 90.63, respectively, highlighting its outstanding performance in coal mine gas risk assessment. Through comparative experiments with other large language models of similar size and manual evaluation, MIRA-ChatGLM demonstrated superior performance across multiple key metrics, fully demonstrating its tremendous potential in intelligent mine risk assessment and decision support.

关键词： intelligent assessment large language models fine-tuning coal mine gas risk quantization algorithm

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：