检索结果-内蒙古大学图书馆

An Optimized encoding scheme for Planning Vessel-Encoded Pseudocontinuous Arterial Spin Labeling

MAGNETIC RESONANCE IN MEDICINE 2015年第5期74卷 1248-1256页

作者： Berry, Eleanor S. K. Jezzard, Peter Okell, Thomas W. Univ Oxford Ctr Funct Magnet Resonance Imaging Brain Nuffield Dept Clin Neurosci Oxford England

PurposeVessel-encoded pseudocontinuous arterial spin labeling allows the acquisition of vessel-selective angiograms and vascular territory perfusion maps. The technique generates a periodic variation in inversion efficiency across space that can be manipulated to encode specific combinations of vessels. Currently, the choice of these encodings is limited to scenarios with few vessels and may not optimize the signal-to-noise ratio (SNR). Here we present an automated, rapid method for calculating a minimal number of SNR optimal encodings for any number and arrangement of vessels. Theory and MethodsThe proposed optimized encoding scheme (OES) is a Fourier-based method that finds SNR optimized encodings to best match the ideal encodings for a set of vessels. For nine or fewer vessels, the calculation takes less than 3 s. ResultsIn simulations, the OES method produces encodings for a range of vessel geometries that, on average, have an SNR efficiency 37% greater than that for random encoding. When labeling vessels in the neck in healthy subjects, the OES encodings result in images with higher SNR than other encoding methods. ConclusionThe OES results in a minimal number of encodings with a higher SNR efficiency than other encoding methods, regardless of the number or geometry of the vessels. Magn Reson Med 74:1248-1256, 2015. (c) 2014 The Authors. Magnetic Resonance in Medicine Published by Wiley Periodicals, Inc. on behalf of International Society of Medicine in Resonance.

关键词： vessel-encoded pseudocontinuous arterial spin labeling vessel-selective encoding scheme SNR efficiency perfusion

来源：评论

学校读者我要写书评

暂无评论

Evaluation of encoding schemes on Ubiquitous Sensor Signal for Spiking Neural Network

引用

IEEE SENSORS JOURNAL 2024年第21期24卷 35008-35018页

作者： Bian, Sizhen Donati, Elisa Magno, Michele Swiss Fed Inst Technol Dept Informat Technol & Elect Engn Ctr Project Based Learning CH-8092 Zurich Switzerland Univ Zurich Inst Neuroinformat CH-8057 Zurich Switzerland Swiss Fed Inst Technol CH-8057 Zurich Switzerland

Spiking neural networks (SNNs), a brain-inspired computing paradigm, are emerging for their inference performance, particularly in terms of energy efficiency and latency attributed to the plasticity in signal processing. To deploy SNNs in ubiquitous computing systems, signal encoding of sensors is crucial for achieving high accuracy and robustness. Using inertial sensor readings for gym activity recognition as a case study, this work comprehensively evaluates four main encoding schemes and deploys the corresponding SNN on the neuromorphic processor Loihi2 for postdeployment encoding assessment. Rate encoding, time-to-first-spike (TTFS) encoding, binary encoding, and delta modulation are evaluated using metrics such as average fire rate, signal-to-noise ratio (SNR), classification accuracy, robustness, and inference latency and energy. In this case study, the TTFS encoding required the lowest firing rate (2%) and achieved a comparative accuracy (89%) although it was the least robust scheme against error spikes (over 20% accuracy drop with 0.1 noisy spike rate). Rate encoding with optimal value-to-probability mapping achieved the highest accuracy (91.7%). Binary encoding provided a balance between information reconstruction and noise resistance. Multithreshold delta modulation showed the best robustness, with only a 0.7% accuracy drop at a 0.1 noisy spike rate. This work serves researchers in selecting the best encoding scheme for SNN-based ubiquitous sensor signal processing, tailored to specific performance requirements.

关键词： encoding scheme neuromorphic computing spiking neural network (SNN) ubiquitous computing

来源：评论

学校读者我要写书评

暂无评论

Does Configuration encoding Matter in Learning Software Performance? An Empirical Study on encoding schemes

Does Configuration Encoding Matter in Learning Software Perf...

引用

19th International Conference on Mining Software Repositories (MSR)

作者： Gong, Jingzhi Chen, Tao Loughborough Univ Loughborough Leics England

ISBN: (纸本)9781450393034

Learning and predicting the performance of a configurable software system helps to provide better quality assurance. One important engineering decision therein is how to encode the configuration into the model built. Despite the presence of different encoding schemes, there is still little understanding of which is better and under what circumstances, as the community often relies on some general beliefs that inform the decision in an ad-hoc manner. To bridge this gap, in this paper, we empirically compared the widely used encoding schemes for software performance learning, namely label, scaled label, and one-hot encoding. The study covers five systems, seven models, and three encoding schemes, leading to 105 cases of investigation. Our key findings reveal that: (1) conducting trial-and-error to find the best encoding scheme in a case by case manner can be rather expensive, requiring up to 400+ hours on some models and systems;(2) the one-hot encoding often leads to the most accurate results while the scaled label encoding is generally weak on accuracy over different models;(3) conversely, the scaled label encoding tends to result in the fastest training time across the models/systems while the one-hot encoding is the slowest;(4) for all models studied, label and scaled label encoding often lead to relatively less biased outcomes between accuracy and training time, but the paired model varies according to the system. We discuss the actionable suggestions derived from our findings, hoping to provide a better understanding of this topic for the community. To promote open science, the data and code of this work can be publicly accessed at https://***/ideas-labo/MSR2022-encoding-study.

关键词： encoding scheme Machine Learning Software Engineering Performance Prediction Performance Learning Configurable Software

来源：评论

学校读者我要写书评

暂无评论

Mission planning for active removal of multiple space debris in low Earth orbit

引用

ADVANCES IN SPACE RESEARCH 2024年第9期73卷 4800-4812页

作者： Choi, Jin Haeng Park, Chandeok Lee, Jinah Yonsei Univ Dept Astron Seoul 03722 South Korea 50 Yonsei Ro Seoul 03722 South Korea

This study presents a novel framework of mission planning for actively removing multiple space debris in low Earth orbit. The associated problem is formulated as a multi -objective optimization problem to minimize the total velocity increments while maximizing the Hazard Criticality Index (HCI) of the candidate target debris. The main idea is based on the Random -Key (RK) encoding scheme that effectively converts a continuous search space into a discretized solution space representing the target debris and visiting sequence. With the RK encoding technique, the proposed framework can obtain the optimal solutions for target debris selection for removal, removal sequences, and other design variables for computing orbital transfer costs simultaneously in a systematic way. In order to demonstrate these favorable characteristics, the proposed framework is integrated with four well-known multi -objective Evolutionary Algorithms (EAs) to solve mission planning problems for the active removal of multiple space debris in low Earth orbit. The framework integrated with NSGA-II, compared with other EAs, shows the best performance in terms of optimality and diversity of non -dominated solutions. Moreover, the optimization results indicate that our integrated framework appropriately selects target debris with similar orbital elements and naturally employs the Right Ascension of the Ascending Node (RAAN) drift originating from J2 perturbation to shift the transfer orbit to reduce total velocity increments. The overall results show that the proposed framework is effective in the preliminary mission planning for multiple space debris removal in low Earth orbit. (c) 2024 COSPAR. Published by Elsevier B.V. All rights reserved.

关键词： Active space debris removal encoding scheme Multi -objective optimization Trajectory optimization

来源：评论

学校读者我要写书评

暂无评论

Speed-Power Efficient Novel CMOS Unary-to-Ternary Encoder

引用

IETE JOURNAL OF RESEARCH 2024年第6期70卷 5896-5904页

作者： Saha, Aloke Sanyal, Prerona Singh, Deep Narayan Bharti, Aastha Pal, Dipankar Dr BC Roy Engn Coll Dept ECE Durgapur India BITS Pilani Dept EEE K K Birla Goa Campus Goa India

A new full-custom design of Unary-to-Ternary-Encoder (UTE) using standard Enhancement-Type-Metal-Oxide-Semiconductor-Field-Effect-Transistor reporting low-PDP (Power-Delay-Product) and low-area is presented. The inherent pass characteristic of the E-MOS-transistor is exploited here to develop the proposed Ternary-Encoder. Theoretical aspects along with the operation of proposed 9:2 and 27:3 UTE are discussed. The complete UTE is designed and optimized on 32 nm standard CMOS technology at 0.9 V supply-rail and 27(degrees)C. Ternary digits "0", "1" and "2" are represented with 0 V, 0.45 V and 0.9 V respectively. The proposed design is validated through extensive T-Spice front-end transient simulations with all possible test patterns. The layout of the proposed design is completed on 32 nm Single-Poly-Double-Metal (SPDM) CMOS--Technology. After DRC and LVS post-layout simulation with extracted parasitic and 1 fF load, is carried out. The valuated performance of the 9:2 UTE is next compared with a recent candidate design to this end, from open literature to benchmark. The Power-Delay-Product (PDP) of proposed UTE is measured by applying +/- 10% supply variation from nominal on slow, typical and fast MOSFET at -40(degrees)C, 27(degrees)C and 85(degrees)C. Finally, the speed-power characteristic of the proposed 9:2 UTE is explored under different load conditions from 1 to 10 fF.

关键词： E-MOS pass characteristic encoding scheme PDP(Power delay product) PVT(Process-voltage-temperature) analyses Transient analyses Ternarysystem

来源：评论

学校读者我要写书评

暂无评论

A multi-disjunctive-graph model-based memetic algorithm for the distributed job shop scheduling problem

引用

ADVANCED ENGINEERING INFORMATICS 2024年 60卷

作者： Wang, Sihan Li, Xinyu Gao, Liang Li, Jiahang Huazhong Univ Sci & Technol State Key Lab Digital Mfg Equipment & Technol Wuhan Peoples R China

With the influence of the digital economy, the traditional manufacturing model is undergoing a shift towards a distributed manufacturing model. This transition involves multiple workshops across diverse geographic regions. The core of distributed manufacturing is the concept of decentralized management and execution, which includes various stages, resources, and tasks within the production process. A key technology in this domain is the distributed shop scheduling problem. Notably, the distributed job shop scheduling problem (DJSSP), considering job shops, is widespread in real distributed manufacturing processes and is difficult to solve as an NP-hard problem. Although several heuristic solvers and metaheuristic algorithms have attempted to address this problem, currently two sub-problems included in the problem, factory allocation and sequence of operations, are treated separately and the description of the problem is incomplete. This paper introduces a multi-disjunctivegraph model-based memetic algorithm (MDGMBMA) developed for DJSSP to minimize the makespan. The multi-disjunctive-graph model is proposed to fully represent the DJSSP solution space. Additionally, an innovative encoding method is proposed to achieve an adequate search of the solution space, and two decoding strategies are proposed to address the search and evaluation demands of the algorithm. Furthermore, based on the property of critical job exchange between factories, a specialized critical job exchange-based neighborhood structure is designed to enhance the efficiency of the tabu search. To evaluate the performance of the MDGMBMA, numerical results from 240 large instances (ranging from 2 to 4 factories) derived from well-known JSSP benchmarks are compared against recently published discrete metaheuristic algorithms. The experimental results indicate that the proposed algorithm performs effectively in solving DJSSP.

关键词： Distributed job shop scheduling problem Multi-disjunctive-graph model Memetic algorithm encoding scheme Neighborhood structure

来源：评论

学校读者我要写书评

暂无评论

FEEBY: A Flexible Framework for Fast Prototyping and Assessment of Vibrotactile Feedback for Hand Prostheses

IEEE TRANSACTIONS ON MEDICAL ROBOTICS AND BIONICS

引用

IEEE TRANSACTIONS ON MEDICAL ROBOTICS AND BIONICS 2024年第2期6卷 746-756页

作者： Maravic, Nikolina Dosen, Strahinja Gasparic, Filip Hofer, Christian Russold, Michael Koppe, Mario Gonzalez-Vargas, Jose Jorgovanovic, Nikola Stanisic, Darko Univ Novi Sad Fac Tech Sci Novi Sad 21102 Serbia Aalborg Univ Dept Hlth Sci & Technol DK-9220 Aalborg Denmark Otto Bock Healthcare Prod GmbH Dept Global Res A-1110 Vienna Austria Ottobock SE & Co KGaA Dept Global Res D-37115 Duderstadt Germany

Enhancing a myoelectric prosthesis with artificial somatosensory feedback is important for providing a complete bionic replacement. The development of feedback comprises several steps, from the selection of stimulation interface, variables, and encoding schemes to testing in non-disabled and amputee subjects. In most cases, specific configurations are implemented and tested. To support a more flexible approach to the development of feedback, where an interface can be iteratively tested and gradually refined, we developed FEEBY - a software and hardware framework for fast prototyping and assessment of feedback both in and out of the lab. FEEBY comprises a PC application for feedback design and subject training, an embedded system for clinical testing, and a smaller version of the system for home use. The system capabilities and the feedback design process were demonstrated by conducting illustrative experiments on 6 non-disabled participants and 1 transradial amputee, who also used the system at home. The results of the present study demonstrated that FEEBY is a compact, low-power, and robust system that can enable the systematic development of feedback and its assessment in ecological conditions. The latter aspect is particularly important to reach the ultimate goal, i.e., a feedback system that matters in daily life.

关键词： encoding scheme embedded system hand prosthesis vibrotactile feedback fast prototyping home use

来源：评论

学校读者我要写书评

暂无评论

Prediction of CRISPR-Cas9 off-target activities with mismatches and indels based on hybrid neural network

引用

COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL 2023年 21卷 5039-5048页

作者： Yang, Yanpeng Li, Jian Zou, Quan Ruan, Yaoping Feng, Hailin Zhejiang A&F Univ Sch Math & Comp Sci Hangzhou 311300 Peoples R China Univ Elect Sci & Technol China Yangtze Delta Reg Inst Quzhou Quzhou 324000 Peoples R China Univ Elect Sci & Technol China Inst Fundamental & Frontier Sci Chengdu 610054 Peoples R China

The CRISPR/Cas9 system has significantly advanced the field of gene editing, yet its clinical application is constrained by the considerable challenge of off-target effects. Although numerous deep learning models for off-target prediction have been proposed, most struggle to effectively extract the nuanced features of guide RNA (gRNA) and DNA sequence pairs and to mitigate information loss during data transmission within the model. To address these limitations, we introduce a novel Hybrid Neural Network (HNN) model that employs a parallelized network structure to fully extract pertinent features from different positions and types of bases in the sequence to minimize information loss. Notably, this study marks the first application of word embedding techniques to extract information from sequence pairs that contain insertions and deletions (Indels). Comprehensive evaluation across diverse datasets indicates that our proposed model outperforms existing state-of-the-art prediction methods in off-target prediction. The datasets and source codes supporting this study can be found at https:// github .com /Yang -k955 /CRISPR -HW.

关键词： CRISPR-Cas9 Off-target prediction Hybrid neural network encoding scheme

来源：评论

学校读者我要写书评

暂无评论

Action Command encoding for Surrogate-Assisted Neural Architecture Search

引用

IEEE TRANSACTIONS ON COGNITIVE AND DEVELOPMENTAL SYSTEMS 2022年第3期14卷 1129-1142页

作者： Tian, Ye Peng, Shichen Yang, Shangshang Zhang, Xingyi Tan, Kay Chen Jin, Yaochu Anhui Univ Inst Phys Sci & Informat Technol Minist Educ Key Lab Intelligent Comp & Signal Proc Hefei 230601 Peoples R China Anhui Univ Sch Comp Sci & Technol Minist Educ Key Lab Intelligent Comp & Signal Proc Hefei 230601 Peoples R China Anhui Univ Sch Artificial Intelligence Minist Educ Key Lab Intelligent Comp & Signal Proc Hefei 230601 Peoples R China Hong Kong Polytech Univ Dept Comp Hong Kong Peoples R China Univ Surrey Dept Comp Sci Guildford GU2 7XH Surrey England

With the development of neural architecture search, the performance of deep neural networks has been considerably enhanced with less human expertise. While the existing work mainly focuses on the development of optimizers, the design of encoding scheme is still in its infancy. This article thus proposes a novel encoding scheme for neural architecture search, termed action command encoding (ACencoding). Inspired by the gene expression process, ACencoding defines several action commands to indicate the addition and clone of layers, connections, and local modules, where an architecture grows from empty according to multiple action commands. ACencoding provides a compact and rich search space that can be explored by various optimizers efficiently. Furthermore, a surrogate-assisted performance evaluator is tailored for ACencoding, termed sequence-to-rank (Seq2Rank). By integrating the Seq2Seq model with RankNet, Seq2Rank embeds the variable-length encoding of ACencoding into a continuous space, and then predicts the rankings of architectures based on the continuous representation. In the experiments, ACencoding brings improvement to neural architecture search with existing encoding schemes and Seq2Rank shows better accuracy than existing performance evaluators. The neural architectures obtained by ACencoding and Seq2Rank have competitive test errors and complexities on image classification tasks, and also show high transferability between different data sets.

关键词： Computer architecture encoding Topology Optimization Evolutionary computation Task analysis Reinforcement learning encoding scheme neural architecture search optimizer performance evaluator

来源：评论

学校读者我要写书评

暂无评论

Effective use of sequence information to predict CRISPR-Cas9 off-target

引用

COMPUTATIONAL AND STRUCTURAL BIOTECHNOLOGY JOURNAL 2022年 20卷 650-661页

作者： Zhang, Zhong-Rui Jiang, Zhen-Ran East China Normal Univ Sch Comp Sci & Technol Shanghai 200062 Peoples R China

The CRISPR/Cas9 gene-editing system is the third-generation gene-editing technology that has been widely used in biomedical applications. However, off-target effects occurring CRISPR/Cas9 system has been a challenging problem it faces in practical applications. Although many predictive models have been developed to predict off-target activities, current models do not effectively use sequence pair information. There is still room for improved accuracy. This study aims to effectively use sequence pair information to improve the model's performance for predicting off-target activities. We propose a new coding scheme for coding sequence pairs and design a new model called CRISPR-IP for predicting off-target activity. Our coding scheme distinguishes regions with different functions in the sequence pairs through the function channel. Moreover, it distinguishes between bases and base pairs using type channels, effectively representing the sequence pair information. The CRISPR-IP model is based on CNN, BiLSTM, and the attention layer to learn features of sequence pairs. We performed performance verification on two data sets and found that our coding scheme can represent sequence pair information effectively, and the CRISPR-IP model performance is better than others. Data and source codes are available at https:// ***/BioinfoVirgo/CRISPR-IP. (c) 2022 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/).

关键词： CRISPR-Cas9 Off-target prediction Deep learning encoding scheme

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：