检索结果-内蒙古大学图书馆

arXiv 2022年

作者： Bai, Lin Zhao, Yiming Huang, Xinming Department of Electrical and Computer Engineering Worcester Polytechnic Institute WorcesterMA01609 United States

Point cloud semantic segmentation has attracted attentions due to its robustness to light condition. This makes it an ideal semantic solution for autonomous driving. However, considering the large computation burden and bandwidth demanding of neural networks, putting all the computing into vehicle Electronic Control Unit (ECU) is not efficient or practical. In this paper, we proposed a light weighted point cloud semantic segmentation network based on range view. Due to its simple pre-processing and standard convolution, it is efficient when running on deep learning accelerator like DPU. Furthermore, a near sensor computing system is built for autonomous vehicles. In this system, a FPGA-based deep learning accelerator core (DPU) is placed next to the LiDAR sensor, to perform point cloud pre-processing and segmentation neural network. By leaving only the post-processing step to ECU, this solution heavily alleviate the computation burden of ECU and consequently shortens the decision making and vehicles reaction latency. Our semantic segmentation network achieved 10 frame per second (fps) on Xilinx DPU with computation efficiency 42.5 GOP/W. Copyright © 2022, The Authors. All rights reserved.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

A Logical Circuit Optimization in Balancing Delay and Energy Consumption 23

A Logical Circuit Optimization in Balancing Delay and Energy...

引用

Proceedings of the 6th International Conference on Information Technologies and electrical engineering

作者： Qihang Shan The Bradley Department of Electrical and Computer Engineering Virginia Polytechnic Institute and State University USA

ISBN: (纸本)9798400708299

The fast-developing chip manufacturing technique and scaling of transistors allow us to fit more transistors on a small chip. The scaling down process, however, is facing a challenge. The smaller transistors are, the more influential quantum channeling and silicon atom size limit become. To improve efficiency, the solution of scaling down is no longer an option. Therefore, to further improve the efficiency of a chip without scaling down transistors, this paper presents a combinational circuit and focuses on an optimization approach where energy consumption is reduced in exchange for increasing delay. By adjusting the size of transistors, energy is saved while maintaining delay to an acceptable range. This approach manages to reduce energy consumption by about 56% while increasing delay by 50%. This paper represents one of many possible approaches that researchers had and has been working on and this tradeoff can benefit some circuit designs depending on the circuit's purpose and hope to bring some insights on further optimization.

关键词： Delay Energy Consumption Optimization Tradeoff

来源：评论

学校读者我要写书评

暂无评论

Improved Bound for Robust Causal Bandits with Linear Models

Improved Bound for Robust Causal Bandits with Linear Models

引用

IEEE International Symposium on Information Theory

作者： Zirui Yan Arpan Mukherjee Burak Varıcı Ali Tajer Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy NY USA

ISBN: (数字)9798350382846

ISBN: (纸本)9798350382853

This paper investigates the robustness of causal bandits (CBs) in the face of temporal model fluctuations. This setting deviates from the existing literature's widely-adopted assumption of constant causal models. The focus is on causal systems with linear structural equation models (SEMs). The SEMs and the time-varying pre- and post-interventional statistical models are all unknown and subject to variations over time. The goal is to design a sequence of interventions that incur the smallest cumulative regret compared to an oracle aware of the entire causal model and its fluctuations. A robust CB algorithm is proposed, and its cumulative regret is analyzed by establishing both upper and lower bounds on the regret. It is shown that in a graph with maximum in-degree $d$ , length of the largest causal path $L$ , and an aggregate model deviation $C$ , the regret is upper bounded by $\tilde{\mathrm{O}}(d^{L-\frac{1}{2}}(\sqrt{T}+C))$ and lower bounded by $\Omega(d^{\frac{L}{2}-2}\max\{\sqrt{T}\,\ d^{2}C\})$ . The proposed algorithm achieves nearly optimal $\tilde{\mathcal{O}}(\sqrt{T})$ regret when $C$ is $o(\sqrt{T})$ , maintaining sub-linear regret for a broad range of C.

关键词： Fluctuations Upper bound Numerical analysis Aggregates Mathematical models Robustness Faces

来源：评论

学校读者我要写书评

暂无评论

BAI in Exponential Family: Efficiency and Optimality

BAI in Exponential Family: Efficiency and Optimality

引用

IEEE International Symposium on Information Theory

作者： Arpan Mukherjee Ali Tajer Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy NY USA

ISBN: (数字)9798350382846

ISBN: (纸本)9798350382853

This paper investigates a hitherto unaddressed as-pect of best arm identification (BAI) in stochastic multi-armed bandits in the fixed -confidence setting. Two essential metrics for assessing bandit algorithms are computational efficiency and performance optimality (e.g., in sample complexity). In stochastic BAI literature, there have been advances in designing algorithms to achieve optimal performance, but they are generally computationally expensive to implement (e.g., optimization-based methods). There also exist approaches that have high computationally efficiency but do not achieve the optimal performance (e.g., UCB-based methods) or achieve it up to a gap (e.g., the $\beta-$ optimal approaches in top-two methods). This paper introduces a framework and an algorithm for BAI that achieves optimal performance with a computationally efficient set of decision rules. The central process that facilitates this is a routine for sequentially estimating the optimal allocations up to sufficient fidelity. Specifically, these estimates are accurate enough for identifying the best arm (hence, achieving optimality) but not excessively accurate to an unnecessary extent (hence, maintaining efficiency). Numerical evaluations are provided to (i) establish the optimality and efficiency of the algorithm, (ii) showcase the implicit estimation property of the proposed allocation rules, and (ii) demonstrate the superior performance of the proposed algorithms compared to the existing ones.

关键词： Measurement Accuracy Costs Estimation Rendering (computer graphics) Computational efficiency Complexity theory

来源：评论

学校读者我要写书评

暂无评论

Graph Coordinates and Conventional Neural Networks - An Alternative for Graph Neural Networks

Graph Coordinates and Conventional Neural Networks - An Alte...

引用

IEEE International Conference on Big Data

作者： Zheyi Qin Randy Paffenroth Anura P. Jayasumana Electrical and Computer Engineering Colorado State University Fort Collins USA Department of Data Science Program Worcester Polytechnic Institute Worcester MA USA

Graph-based data present unique challenges and opportunities for machine learning. Graph Neural Networks (GNNs), and especially those algorithms that capture graph topology through message passing for neighborhood aggregation, have been a leading solution. However, these networks often require substantial computational resources and may not optimally leverage the information contained in the graph’s topology, particularly for large-scale or complex *** propose Topology Coordinate Neural Network (TCNN) and Directional Virtual Coordinate Neural Network (DVCNN) as novel and efficient alternatives to message passing GNNs, that directly leverage the graph’s topology, sidestepping the computational challenges presented by competing algorithms. Our proposed methods can be viewed as a reprise of classic techniques for graph embedding for neural network feature engineering, but they are novel in that our embedding techniques leverage ideas in Graph Coordinates (GC) that are lacking in current *** results, benchmarked against the Open Graph Benchmark Leaderboard, demonstrate that TCNN and DVCNN achieve competitive or superior performance to message passing GNNs. For similar levels of accuracy and ROC-AUC, TCNN and DVCNN need far fewer trainable parameters than contenders of the OGBN Leaderboard. The proposed TCNN architecture requires fewer parameters than any neural network method currently listed in the OGBN Leaderboard for both OGBN-Proteins and OGBN-Products datasets. Conversely, our methods achieve higher performance for a similar number of trainable parameters. These results hold across diverse datasets and edge features, underscoring the robustness and generalizability of our methods. By providing an efficient and effective alternative to message passing GNNs, our work expands the toolbox of techniques for graph-based machine learning. A significantly lower number of tunable parameters for a given evaluation metric makes TCNN and DVCNN especiall

关键词：

来源：评论

学校读者我要写书评

暂无评论

Information-Exchange-Reduced Distributed FixedTime Secondary Control for DC Microgrid with Enhanced Stability

引用

IEEE Transactions on Industry Applications 2025年

作者： Chai, Junwei Xu, Zhao Wang, Minghao Lyu, Xue The Hong Kong Polytechnic University Department of Electrical and Electronics Engineering Hong Kong The Hong Kong Polytechnic University Department of Electrical and Electronic Engineering Shenzhen Research Institute Research Institute of Smart Energy Hong Kong Department of Electrical and Computer Engineering Zhuhai China University of Macau Advanced Research Institute Hengqin Zhuhai China

DC microgrids are a feasible grid system to host multiple distributed generators (DGs). In terms of the control and coordination of hosted DGs, the enhanced stability and fast dynamics of DC microgrids are required. To achieve these objectives, this paper proposes a comprehensive strategy that meticulously accounts for the impacts of transmission line impedances and constant power loads (CPLs). The strategy is architected with a dual-layer control approach: a stability enhancement controller is embedded within the primary control layer, while a novel distributed fixed-time controller is deployed in the secondary control layer, ensuring a robust and fast system performance. Specifically, for the primary control layer, the maximum tolerable CPL is firstly calculated using the reduced-order model and the impedance ratio criterion, contributing to more accurate and less conservative results. Additionally, a stability enhancement controller is designed to reshape the system's equivalent output impedance, improving both steady-state and transient stability of the DC microgrid and enabling the accommodation of larger CPLs. For the secondary control layer, a novel fixed-time secondary control scheme is put forward to achieve a concurrent DC bus voltage regulation and current sharing among DGs within a fixed time. Importantly, this method requires the transmission of only one variable to neighboring DGs, significantly decreasing communication traffic compared to the existing latest methods. The stability of the proposed scheme is theoretically proved through Lyapunov-based analysis. Case study results are presented to demonstrate the efficacy of the proposed strategy. © 1972-2012 IEEE.

关键词： Publishing

来源：评论

学校读者我要写书评

暂无评论

High-Fidelity Document Stain Removal via A Large-Scale Real-World Dataset and A Memory-Augmented Transformer

High-Fidelity Document Stain Removal via A Large-Scale Real-...

引用

2025 IEEE/CVF Winter Conference on Applications of computer Vision, WACV 2025

作者： Li, Mingxian Sun, Hao Lei, Yingtie Zhang, Xiaofeng Dong, Yihang Zhou, Yilin Li, Zimeng Chen, Xuhang School of Computer Science and Engineering Huizhou University China The Ohio State University Department of Computer Science and Engineering United States School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University China Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China School of Electronic and Communication Engineering Shenzhen Polytechnic University China

ISBN: (纸本)9798331510831

Document images are often degraded by various stains, significantly impacting their readability and hindering downstream applications such as document digitization and analysis. The absence of a comprehensive stained document dataset has limited the effectiveness of existing document enhancement methods in removing stains while preserving fine-grained details. To address this challenge, we construct StainDoc, the first large-scale, high-resolution (2145 x 2245) dataset specifically designed for document stain removal. StainDoc comprises over 5,000 pairs of stained and clean document images across multiple scenes. This dataset encompasses a diverse range of stain types, severities, and document backgrounds, facilitating robust training and evaluation of document stain removal algorithms. Furthermore, we propose StainRestorer, a Transformer-based document stain removal approach. StainRestorer employs a memory-augmented Transformer architecture that captures hierarchical stain representations at part, instance, and semantic levels via the DocMemory module. The Stain Removal Transformer (SRTransformer) leverages these feature representations through a dual attention mechanism: an enhanced spatial attention with an expanded receptive field, and a channel attention captures channel-wise feature importance. This combination enables precise stain removal while preserving document content integrity. Extensive experiments demonstrate StainRestorer's superior performance over state-of-the-art methods on the Stain-Doc dataset and its variants *** and ***, establishing a new benchmark for document stain removal. Our work highlights the potential of memory-augmented Transformers for this task and contributes a valuable dataset to advance future research. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Leveraging Large Language Models for Wireless Symbol Detection via In-Context Learning

Leveraging Large Language Models for Wireless Symbol Detecti...

引用

IEEE Conference on Global Communications (GLOBECOM)

作者： Momin Abbas Koushik Kar Tianyi Chen Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy NY USA

ISBN: (数字)9798350351255

ISBN: (纸本)9798350351262

Deep neural networks (DNNs) have made significant strides in tackling challenging tasks in wireless systems, especially when an accurate wireless model is not available. However, when available data is limited, traditional DNNs often yield subpar results due to underfitting. At the same time, large language models (LLMs) exemplified by GPT-3, have remarkably showcased their capabilities across a broad range of natural language processing *** whether and how LLMs can benefit challenging non-language tasks in wireless systems is *** this work, we propose to leverage the in-context learning ability (a.k.a. prompting) of LLMs to solve wireless tasks in the low data regime without any training or fine-tuning, unlike DNNs which require training. We further demonstrate that the performance of LLMs varies significantly when employed with different prompt templates. To solve this issue, we employ the latest LLM calibration methods. Our results reveal that using LLMs via ICL methods generally outperforms traditional DNNs on the symbol demodulation task and yields highly confident predictions when coupled with calibration techniques.

关键词： Wireless communication Training Uncertainty Large language models Symbols Artificial neural networks Linguistics Natural language processing Calibration Demodulation

来源：评论

学校读者我要写书评

暂无评论

Thermally robust compact single Bragg grating silicon modulator for photonic computing

引用

Applied Optics 2025年第16期64卷 4646-4653页

作者： Meng Zhang Nicholas Gangi Amir Begović Alexer Chen Bowen Liu Z Rena Huang Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute Troy New York 12180 USA Department of Physics Applied Physics and Astronomy Rensselaer Polytechnic Institute Troy New York 12180 USA

We report, to our knowledge, a novel compact, thermally operated optical modulator with 6-bit resolution. The device comprises a slow-light photonic structure, a Bragg grating with a footprint of $10\,\,{\unicode{x00B5}{\rm m}}$ by $50\,\,{\unicode{x00B5}{\rm m}}$ . This single Bragg-grating modulator (SBG-M) operates based on thermally induced band edge shift of the photonic stop band and demonstrates almost constant normalized mean square error (NMSE) lower than 0.001 from room temperature to ${40^\circ}{\rm C}$ , while maintaining thermal robustness up to ${65^\circ}{\rm C}$ limited by the Johnson noise of the on-chip photodetector. The clock frequency, computing accuracy, power consumption, thermal stability, and fabrication-induced spectrum non-ideality are thoroughly studied. The SBG-M achieves a measured lowest NMSE of $5.0 \times {10^{- 4}}$ , with an average power consumption of 5.29 mW. Compared to other optical devices such as micro-ring resonators and slow-light incorporated Mach–Zehnder modulators, the reported device demonstrates significant advantages in compactness, lower power consumption, and thermal robustness.

关键词： Optical computing Optical modulators Phase shift Quantum computation Silicon modulators Silicon photonics

来源：评论

学校读者我要写书评

暂无评论

Exploring Convolutional Neural Networks for Fingerprint Damage Detection and Classification

Exploring Convolutional Neural Networks for Fingerprint Dama...

引用

2024 International Conference on Intelligent Systems for Cybersecurity, ISCS 2024

作者： Swathi, R. Kansal, Vipashi Shyamala Valsalan, Prajoona Vekariya, Vipul Taqui, Syed Noeman Prasad v Potluri Siddhartha Institute of Technology Department of Electrical and Electronics Engineering Andhra Pradesh Vijayawada521108 India Graphic Era Deemed to Be University Department of Computer Science and Engineering Uttarakhand Dehradun248002 India Government Polytechnic for Women Department of Electronics and Communication Engineering Karnataka Ramannagara562159 India Dhofar University Department of Electrical and Computer Engineering Dhofar Salalah Oman Parul Institute of Engineering and Technology Parul University Department of Computer Science and Engineering Post Limda Gujarat Waghodia391760 India Saveetha School of Engineering Simts Saveetha University Department of Vlsi Microelectronics Tamil Nadu Chennai602105 India

ISBN: (纸本)9798350375237

This study proposes a novel approach to fingerprint classification using Deep Convolutional Neural Networks (CNNs), leveraging datasets from prominent sources like the National institute of Standards and Technology (NIST), Fingerprint Verification Competition (FVC), and the SHUFFLE database. The research involves a systematic pipeline, beginning with data collection and preprocessing to create a comprehensive dataset that represents real-world variations in fingerprint patterns. A tailored CNN architecture is designed to capture hierarchical features essential for accurate classification, trained on the amalgamated dataset using state-of-the-art optimization algorithms and loss functions. The model's performance is rigorously evaluated across diverse fingerprint patterns using separate datasets from NIST, FVC, and SHUFFLE, showcasing its superiority and adaptability compared to existing benchmark methods. Additionally, the study addresses privacy and ethical considerations associated with biometric data usage, ensuring the model's reliability and applicability in various fingerprint recognition applications. © 2024 IEEE.

关键词： Database systems

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：