检索结果-内蒙古大学图书馆

24th IEEE International Workshop on Signal processing Advances in Wireless Communications, SPAWC 2023

作者： Dai, Weijie Liu, Junkai Wang, Rui Jiang, Yi Fudan University School of Information Science and Technology Department of Communication Science and Engineering Shanghai China

ISBN: (纸本)9781665496261

In this paper, we present a novel method of downlink precoding for cell-free massive multiple-input multiple-output (MIMO) systems using over-the-air (OTA) training. By drawing analogies between a cell-free massive MIMO system and an artificial neural network (ANN), we borrow the idea of back-propagation algorithm to optimize the precoders and combiners via OTA signal exchanges, without incurring channel state information (CSI) estimation or CSI aggregation over some backhaul lines. Numerical simulations show that our method outperforms the state-of-the-art methods in average sum-rate, is robust against pilot contamination, and has lower computational complexity. © 2023 IEEE.

关键词： Channel state information

来源：评论

学校读者我要写书评

暂无评论

Training A Deep Reinforcement Learning Agent for Microgrid Control using PSCAD Environment

Training A Deep Reinforcement Learning Agent for Microgrid C...

引用

IEEE PES Grid Edge Technologies Conference and Exposition (Grid Edge)

作者： Soofi, Arash Farokhi Bayani, Reza Yazdanibiouki, Mehrdad Manshadi, Saeed D. Univ Calif San Diego Elect & Comp Engn San Diego CA 92103 USA San Diego State Univ Elect & Comp Engn San Diego CA 92182 USA

ISBN: (纸本)9781665460125

The accessibility of real-time operational data along with breakthroughs in processing power have promoted the use of Machine Learning (ML) applications in current power systems. Prediction of device failures, meteorological data, system outages, and demand are among the applications of ML in the electricity grid. In this paper, a Reinforcement Learning (RL) method is utilized to design an efficient energy management system for grid-tied Energy Storage Systems (ESS). We implement a Deep Q-Learning (DQL) approach using Artificial neural networks (ANN) to design a microgrid controller system simulated in the PSCAD environment. The proposed on-grid controller coordinates the main grid, aggregated loads, renewable generations, and Advanced Energy Storage (AES). To reduce the cost of operating AESs, the designed controller takes the hourly energy market price into account in addition to physical system characteristics.

关键词： Microgrid energy management system distributed energy resources Reinforcement learning Deep Q-learning Artificial neural network

来源：评论

学校读者我要写书评

暂无评论

ML-based Translation Methods for Protocols and Data Formats 19

ML-based Translation Methods for Protocols and Data Formats

引用

19th International Conference on network and Service Management (CNSM) - network and Service Management in the Era of Generative AI and Digital Twins

作者： Tothfalusi, Tamas Varga, Eszter Csiszar, Zoltan Varga, Pal AITIA Int Inc Ind IoT Div Budapest Hungary Budapest Univ Technol & Econ Dept Telecommun & Media Informat Budapest Hungary

ISBN: (纸本)9783903176591

In order to exchange information between systems, the information must get encoded into a predefined data format, and it must be transferred in a protocol that the communicating parties have agreed upon. This works well if all parties follow the same protocol standard and use the same data description schemes. If systems use different data formats or protocols, then some sort of translation is required. Protocol and data format translation has been attempted previously through rule-based approaches, ontologies, and also by using machine learning (ML) techniques. Due to the current advances related to AI/ML methods, tools, and infrastructure, the accuracy and feasibility of "translation" with ML-approaches improved significantly. This paper introduces a generic approach and methodology for translating data formats and protocols with ML-based methods and presents our initial results through JSON-XML and JSON-SenML translation.

关键词： protocol translation machine learning neural machine translation natural language processing LLM

来源：评论

学校读者我要写书评

暂无评论

Model-Architecture Co-Design for High Performance Temporal GNN Inference on FPGA 36

Model-Architecture Co-Design for High Performance Temporal G...

引用

36th IEEE International Parallel and distributed processing Symposium (IEEE IPDPS)

作者： Zhou, Hongkuan Zhang, Bingyi Kannan, Rajgopal Prasanna, Viktor Busart, Carl Univ Southern Calif Los Angeles CA 90007 USA US Army Res Lab Adelphi MD USA

ISBN: (纸本)9781665481069

Temporal Graph neural networks (TGNNs) are powerful models to capture temporal, structural, and contextual information on temporal graphs. The generated temporal node embeddings outperform other methods in many downstream tasks. Real-world applications require high performance inference on real-time streaming dynamic graphs. However, these models usually rely on complex attention mechanisms to capture relationships between temporal neighbors. In addition, maintaining vertex memory suffers from intrinsic temporal data dependency that hinders task-level parallelism, making it inefficient on general-purpose processors. In this work, we present a novel model-architecture co-design for inference in memory-based TGNNs on FPGAs. The key modeling optimizations we propose include a light-weight method to compute attention scores and a related temporal neighbor pruning strategy to further reduce computation and memory accesses. These are holistically coupled with key hardware optimizations that leverage FPGA hardware. We replace the temporal sampler with an on-chip FIFO based hardware sampler and the time encoder with a look-up-table. We train our simplified models using knowledge distillation to ensure similar accuracy vis- ' a-vis the original model. Taking advantage of the model optimizations, we propose a principled hardware architecture using batching, pipelining, and prefetching techniques to further improve the performance. We also propose a hardware mechanism to ensure the chronological vertex updating without sacrificing the computation parallelism. We evaluate the performance of the proposed hardware accelerator on three real-world datasets. The proposed model reduces the computation complexity by 84% and memory accesses by 67% with less than 0.33% accuracy loss. Compared with CPU/GPU, our FPGA accelerator achieves 16.4/2.3x speedup in latency and 0.27% improvement in accuracy compared with the state-of-the-art inference algorithm. To the best of our knowledge, this

关键词： Temporal Graph neural network Hardware Architecture FPGA

来源：评论

学校读者我要写书评

暂无评论

Development of a CNN-based hierarchical inspection system for detecting defects on electroluminescence images of single-crystal silicon photovoltaic modules

引用

MATERIALS TODAY COMMUNICATIONS 2022年 31卷

作者： Lin, Keh-Moh Lin, Horng-Horng Lin, You-Teh Southern Taiwan Univ Sci & Technol Dept Mech Engn Tainan Taiwan Southern Taiwan Univ Sci & Technol Dept Comp Sci & Informat Engn Tainan Taiwan

Electroluminescence (EL) image technology has long been a standard tool for detecting defects in the photo-voltaic (PV) industry. In recent years, deep learning technology has been developed rapidly and applied to in-line inspection of EL images of PV modules, but not without challenges. In addition to imbalanced data, the recall rate and inspection speed for detecting defects of PV modules are unable to meet industrial expectations. Therefore, this study proposes a hierarchical inspection system to address these challenges. To alleviate the problem of imbalanced data, an EL image of sc-Si solar modules is cropped into images of single solar cells. This allows us to reduce the processing size of a large EL image without overlooking small defects and to increase the amount of data samples for learning fine-to-coarse defect features. To speed up detection, we design two light-weighted convolutional neural network (CNN) models according to feature map analysis. Experimental results show the precision and recall rate of our CNN models on the testing dataset reach 99.36% and 98.77% respectively, which are confirmed by t-distributed stochastic neighbor embedding visualization. With the same detection discrimi-nability, our model needs only 49.63% of the detection time of ResNet-50. After alleviating the problem of imbalanced data, the performance of this hierarchical inspection system meets the requirement for the in-line inspection of the industry.

关键词： Electroluminescence Photovoltaic module Defect Convolutional neural network Deep learning Hierarchical system

来源：评论

学校读者我要写书评

暂无评论

Convolutional neural network for Printed circuit Board Verification 2

Convolutional Neural Network for Printed circuit Board Verif...

引用

2nd IEEE International Conference on distributed Computing and Electrical Circuits and Electronics, ICDCECE 2023

作者： Snehith, Namgiri Bodempudi, Swarna Sai, K. Rohith Satya Vardhani, M. Shameem, Syed Anudeep, S.D.V. Sai Koneru Lakshmaiah Education Foundation Department of Ece Guntur District Vaddeswaram India

ISBN: (纸本)9798350347456

One of the most important steps in the manufacturing of electronic goods is the creation of printed circuit boards (PCB). An electronic device's PCBs are the first stage in production, so even a small error there could cause serious flaws in the finished product. Finding and identifying flaws in PCBs is crucial given the size and demand of the PCB industry. In this paper, we propose a reference comparison method for PCB defect localization and classification that employs a template and a test picture and integrates machine learning and image processing techniques. We introduce a solution pipeline that works in three stages: Subtraction of Images where the test images are subtracted to generate a mask that highlights the regions of defect on the test image, Contour search to find the contour of the defects from the generated test image, and Model Inference where the contours found above are extracted from the images and fed into our machine learning model which then classifies those defects in the test image. We train the machine learning model, in our case, a Convolutional neural network (CNN) that forms the artificially intelligent (AI) part of our solution which can identify the type of defect present given the image of the defect as input. Our experimental findings demonstrate that, compared to other traditional methods utilized for this work, our model diagnoses the afflicted parts with 97% accuracy. © 2023 IEEE.

关键词： Printed circuit boards

来源：评论

学校读者我要写书评

暂无评论

Explainable neural networks that simulate reasoning

NATURE COMPUTATIONAL SCIENCE

引用

NATURE COMPUTATIONAL SCIENCE 2021年第9期1卷 607-618页

作者： Blazek, Paul J. Lin, Milo M. Univ Texas Southwestern Med Ctr Dallas Green Ctr Syst Biol Dallas TX 75390 USA Univ Texas Southwestern Med Ctr Dallas Dept Bioinformat Dallas TX 75390 USA Univ Texas Southwestern Med Ctr Dallas Dept Biophys Dallas TX 75390 USA Univ Texas Southwestern Med Ctr Dallas Ctr Alzheimers & Neurodegenerat Dis Dallas TX 75390 USA

The success of deep neural networks suggests that cognition may emerge from indecipherable patterns of distributed neural activity. Yet these networks are pattern-matching black boxes that cannot simulate higher cognitive functions and lack numerous neurobiological features. Accordingly, they are currently insufficient computational models for understanding neural information processing. Here, we show how neural circuits can directly encode cognitive processes via simple neurobiological principles. To illustrate, we implemented this model in a non-gradient-based machine learning algorithm to train deep neural networks called essence neural networks (ENNs). neural information processing in ENNs is intrinsically explainable, even on benchmark computer vision tasks. ENNs can also simulate higher cognitive functions such as deliberation, symbolic reasoning and out-of-distribution generalization. ENNs display network properties associated with the brain, such as modularity, distributed and localist firing, and adversarial robustness. ENNs establish a broad computational framework to decipher the neural basis of cognition and pursue artificial general intelligence.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Detecting Ethereum Phishing Scams via Multi-transaction-view Graph Attention network 29

Detecting Ethereum Phishing Scams via Multi-transaction-view...

引用

29th IEEE International Conference on Parallel and distributed Systems, ICPADS 2023

作者： Zhuo, Shaoxuan Li, Guang Wu, Weigang Bian, Jing Sun Yat-Sen University School of Computer Science and Engineering Guangzhou China Guangdong Province Key Laboratory of Big Data Analysis and Processing China Sun Yat-Sen University Guangdong Province Key Laboratory of Computational Science Guangzhou China

ISBN: (纸本)9798350330717

Cryptocurrency phishing scams is a significant treat to Ethereum, one of the most popular blockchain platforms. Most of existing Ethereum phishing detection methods are based on traditional machine learning or graph representation learning, which mostly rely on only statistical and structural features in local scope. In this paper, we propose Multi-transaction-view Graph Attention network (MTvGAT), a novel phishing scam detection model that can make use of transaction patterns of different scopes. To obtain global-view information, we apply graph clustering and construct the global-view graph with multiple clusters, including all the nodes of the original transaction network. To obtain local view information, we apply neighborhood sampling, and construct local-view graphs with target nodes and their neighborhood nodes. Then, node features, edge features, and attention coefficients are aggregated to merge multi-view information into representation of nodes. We further combine global-view and local-view representations to finally identify phishing addresses from target nodes. Extensive experiments demonstrate that the proposed method can outperform existing ones with significant improvement. © 2023 IEEE.

关键词： Ethereum Graph neural network Phishing scams

来源：评论

学校读者我要写书评

暂无评论

Decentralized and Model-Free Federated Learning: Consensus-Based Distillation in Function Space

引用

IEEE TRANSACTIONS ON SIGNAL AND INFORMATION processing OVER networkS 2022年 8卷 799-814页

作者： Taya, Akihito Nishio, Takayuki Morikura, Masahiro Yamamoto, Koji Aoyama Gakuin Univ Coll Sci & Engn Dept Integrated Informat Technol Sagamihara Kanagawa 2525258 Japan Tokyo Tech Sch Engn Meguro 1528550 Japan Kyoto Univ Grad Sch Informat Kyoto 6068501 Japan

This paper proposes a fully decentralized federated learning (FL) scheme for Internet of Everything (IoE) devices that are connected via multi-hop networks. Because FL algorithms hardly converge the parameters of machine learning (ML) models, this paper focuses on the convergence of ML models in function spaces. Considering that the representative loss functions of ML tasks e.g., mean squared error (MSE) and Kullback-Leibler (KL) divergence, are convex functionals, algorithms that directly update functions in function spaces could converge to the optimal solution. The key concept of this paper is to tailor a consensus-based optimization algorithm to work in the function space and achieve the global optimum in a distributed manner. This paper first analyzes the convergence of the proposed algorithm in a function space, which is referred to as a meta-algorithm, and shows that the spectral graph theory can be applied to the function space in a manner similar to that of numerical vectors. Then, consensus-based multi-hop federated distillation (CMFD) is developed for a neural network (NN) to implement the meta-algorithm. CMFD leverages knowledge distillation to realize function aggregation among adjacent devices without parameter averaging. An advantage of CMFD is that it works even with different NN models among the distributed learners. Although CMFD does not perfectly reflect the behavior of the meta-algorithm, the discussion of the meta-algorithm's convergence property promotes an intuitive understanding of CMFD, and simulation evaluations show that NN models converge using CMFD for several tasks. The simulation results also show that CMFD achieves higher accuracy than parameter aggregation for weakly connected networks, and CMFD is more stable than parameter aggregation methods.

关键词： Consensus-based distributed optimization distributed learning federated learning IoE knowledge distillation multi-hop network machine learning

来源：评论

学校读者我要写书评

暂无评论

Stash: A comprehensive stall-centric characterization of public cloud VMs for distributed deep learning 43

Stash: A comprehensive stall-centric characterization of pub...

引用

43rd IEEE International Conference on distributed Computing Systems (ICDCS)

作者： Sharma, Aakash Bhasi, Vivek M. Singh, Sonali Jain, Rishabh Gunasekaran, Jashwant Raj Mitra, Subrata Kandemir, Mahmut Taylan Kesidis, George Das, Chita R. Penn State Univ Comp Sci & Engn University Pk PA 16802 USA Adobe Res San Francisco CA USA

ISBN: (纸本)9798350339864

Deep neural networks (DNNs) are increasingly popular owing to their ability to solve complex problems such as image recognition, autonomous driving, and natural language processing. Their growing complexity coupled with the use of larger volumes of training data (to achieve acceptable accuracy) has warranted the use of GPUs and other accelerators. Such accelerators are typically expensive, with users having to pay a high upfront cost to acquire them. For infrequent use, users can, instead, leverage the public cloud to mitigate the high acquisition cost. However, with the wide diversity of hardware instances (particularly GPU instances) available in public cloud, it becomes challenging for a user to make an appropriate choice from a cost/performance standpoint. In this work, we try to address this problem by (i) introducing a comprehensive distributed deep learning (DDL) profiler Stash, which determines the various execution stalls that DDL suffers from, and (ii) using Stash to extensively characterize various public cloud GPU instances by running popular DNN models on them. Specifically, it estimates two types of communication stalls, namely, interconnect and network stalls, that play a dominant role in DDL execution time. Stash is implemented on top of prior work, DS-analyzer, that computes only the CPU and disk stalls. Using our detailed stall characterization, we list the advantages and shortcomings of public cloud GPU instances for users to help them make an informed decision(s). Our characterization results indicate that the more expensive GPU instances may not be the most performant for all DNN models and that AWS can sometimes sub-optimally allocate hardware interconnect resources. Specifically, the intra-machine interconnect can introduce communication overheads of up to 90% of DNN training time and the network-connected instances can suffer from up to 5x slowdown compared to training on a single instance. Furthermore, (iii) we also model the impact of DNN m

关键词： Characterization Cloud computing distributed deep learning Public cloud GPU

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：