检索结果-内蒙古大学图书馆

2023 International Conference on Neuromorphic Computing, ICNC 2023

作者： Wu, Kaiyue Huang, Changwu Yao, Xin Southern University of Science and Technology Guangdong Provincial Key Laboratory of Brain-inspired Intelligent Computation Department of Computer Science and Engineering Shenzhen518055 China Southern University of Science and Technology Research Institute of Trustworthy Autonomous Systems Shenzhen518055 China

ISBN: (纸本)9798350316889

Estimating the confidence of a machine learning (ML) model plays an important role in minimizing undesirable outcomes and safety risks when using ML models in the real world, especially in high-stake application scenarios. However, the majority of existing confidence estimation methods do not work well without incorporating internal information from the model, such as the gradients, internal representations or checkpoints of its training process, which might compromise the security of the model. To alleviate such limitations, we intend to design a reliable confidence estimation method that requires as little internal model information as possible, yet precisely detecting uncertain samples for the model. In this work, we propose a confidence estimation approach that learns the confidence criterion (that is, the true class probability, TCP) based on the explanations of model's predictions. The proposed approach utilizes the knowledge learned in the model with the leverage of the explanations for model's prediction rather than directly extracting internal information of the model, so that it can be adopted in a wide range of applications. Several experiments, including experiments on different datasets, different models and non-IID (independent and identically distributed) test datasets, are conducted to examine the effectiveness of our confidence estimation method. The experimental results show that our proposed approach provides a reliable confidence estimation and generally performs better than the compared methods, especially in the capability of recognizing when a ML model makes errors. © 2023 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

A metasurface-based light-to-microwave transmitter for hybrid wireless communications

引用

Light(science & Applications) 2022年第6期11卷 1139-1148页

作者： Xin Ge Zhang Ya Lun Sun Bingcheng Zhu Wei Xiang Jiang Qian Yu Han Wei Tian Cheng-Wei Qiu Zaichen Zhang Tie Jun Cui State Key Laboratory of Millimeter Waves School of Information Science and EngineeringSoutheast University210096 NanjingChina National Mobile Communications Research Laboratory School of Information Science and EngineeringSoutheast University210096 NanjingChina Frontiers Science Center for Mobile Information Communication and Security Southeast University210096 NanjingChina Purple Mountain Laboratories 211111 NanjingChina Department of Electrical and Computer Engineering National University of SingaporeSingapore 117583Singapore

Signal conversion plays an important role in many applications such as communication,sensing,and *** signal conversion between optical and microwave frequencies is a crucial step to construct hybrid communication systems that combine both optical and microwave wireless technologies to achieve better features,which are highly desirable in the future wireless ***,such a signal conversion process typically requires a complicated relay to perform multiple operations,which will consume additional hardware/time/energy ***,we report a light-to-microwave transmitter based on the time-varying and programmable metasurface integrated with a high-speed photoelectric detection circuit into a *** a transmitter can convert a light intensity signal to two microwave binary frequency shift keying signals by using the dispersion characteristics of the metasurface to implement the frequency division *** illustrate the metasurface-based transmitter,a hybrid wireless communication system that allows dual-channel data transmissions in a light-to-microwave link is demonstrated,and the experimental results show that two different videos can be transmitted and received simultaneously and *** metasurface-enabled signal conversion solution may enrich the functionalities of metasurfaces,and could also stimulate new information-oriented applications.

关键词： microwave surface light

来源：评论

学校读者我要写书评

暂无评论

Counterexample Guided Inductive Synthesis Using Large Language Models and Satisfiability Solving

Counterexample Guided Inductive Synthesis Using Large Langua...

引用

MILCOM, Military Communications Conference

作者： Sumit Kumar Jha Susmit Jha Patrick Lincoln Nathaniel D. Bastian Alvaro Velasquez Rickard Ewetz Sandeep Neema Computer Science Department Florida International University Computer Science Laboratory SRI International Army Cyber Institute United States Military Academy Department of Computer Science University of Colorado Boulder Electrical and Computer Engineering University of Central Florida Electrical and Computer Engineering Vanderbilt University

Generative large language models (LLMs) can follow human-provided instruction prompts and generate human-like responses. Apart from natural language responses, they have been found to be effective at generating formal artifacts such as code, plans, and logical specifications. Despite their remarkably improved accuracy, these models are still known to produce factually incorrect or contextually inappropriate results despite their syntactic coherence – a phenomenon often referred to as hallucinations. This limitation makes it difficult to use these models to synthesize formal artifacts used in safety-critical applications. Unlike tasks such as text summarization and question-answering, bugs in code, plan, and other formal artifacts produced by LLMs can be catastrophic. We posit that we can use the satisfiability modulo theory (SMT) solvers as deductive reasoning engines to analyze the generated solutions from the LLMs, produce counterexamples when the solutions are incorrect, and provide that feedback to the LLMs exploiting the dialog capability of LLMs. This interaction between inductive LLMs and deductive SMT solvers can iteratively steer the LLM to generate the correct response. In our experiments, we use planning over the domain of blocks as our synthesis task for evaluating our approach. We use GPT-4, GPT3.5 Turbo, Davinci, Curie, Babbage, and Ada as the LLMs and Z3 as the SMT solver. Our method allows the user to communicate the planning problem in natural language; even the formulation of queries to SMT solvers is automatically generated from natural language. Thus, the proposed technique can enable non-expert users to describe their problems in natural language, and the combination of LLMs and SMT solvers can produce provably correct solutions.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient displacement convex optimization with particle gradient descent

arXiv

引用

arXiv 2023年

作者： Daneshmand, Hadi Lee, Jason D. Jin, Chi Laboratory for Information and Decision Systems MIT United States Department of Electrical Engineering and Computer Science Princeton Universit United States

Particle gradient descent, which uses particles to represent a probability measure and performs gradient descent on particles in parallel, is widely used to optimize functions of probability measures. This paper considers particle gradient descent with a finite number of particles and establishes its theoretical guarantees to optimize functions that are displacement convex in measures. Concretely, for Lipschitz displacement convex functions defined on probability over Rd, we prove that O(1/ϵ2) particles and O(d/ϵ4) computations are sufficient to find the ϵ-optimal solutions. We further provide improved complexity bounds for optimizing smooth displacement convex functions. We demonstrate the application of our results for function approximation with specific neural architectures with two-dimensional inputs. © 2023, CC BY-NC-ND.

关键词： Convex optimization

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Adaptive Filter for Head Tracking in Augmented Reality (AR)-based Flight Simulators

引用

IEEE Transactions on computers 2025年

作者： Nwobodo, Onyeka J. Kuaban, Godlove Suila Nkemeni, Valery Wereszczynski, Kamil Cyran, Krzysztof A. Silesian University of Technology Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Gliwice Poland Polish Academy of Sciences Institute of Theoretical and Applied Informatics Poland University of Buea Laboratory of Electrical Engineering and Computing Faculty of Engineering and Technology Buea Cameroon

This paper presents a novel approach for head tracking in augmented reality (AR) flight simulators using an adaptive fusion of Kalman and particle filters. This fusion dynamically balances the strengths of both algorithms, leveraging Kalman filters for computational efficiency and particle filters for handling non-linearities based on real-time factors such as sensor noise and head movement patterns. Our method demonstrates superior tracking precision and reduced latency, making it particularly effective for immersive pilot training in AR-based flight simulations. While focused on flight simulation, our approach holds high potential for broader applications in other AR and virtual reality (VR) environments where precise, real-time head tracking is crucial. These results provide actionable design guidance for developers optimizing tracking systems in environments that require fast response times and high accuracy. Future extensions of this research could explore the generalization of our approach across diverse AR applications, including human-computer interaction (HCI), medical simulations, and gaming. © 1968-2012 IEEE.

关键词： Flight simulators

来源：评论

学校读者我要写书评

暂无评论

A Robust Polyscale Length Complexity Measure for Stochastic Self-Affine Processes 20

A Robust Polyscale Length Complexity Measure for Stochastic ...

引用

20th IEEE International Conference on Cognitive Informatics and Cognitive Computing, ICCI*CC 2021

作者： Kinsner, Witold University of Manitoba Cognitive Systems Laboratory Department of Electrical and Computer Engineering Winnipeg Canada

ISBN: (纸本)9781665421195

Cognitive and other nonlinear systems often involve deterministic differentiable processes and stochastic non-differentiable processes. Measuring the complexity of such processes is important when extracting objective features from the processes for their classification in either reactive, or adaptive, or predictive control. This applies to classifiers based not only to the traditional neural networks, but also to deep learning systems, and particularly in cognitive systems. This paper describes a robust algorithm to measure the length complexity of a self-Affine time series using multiscale and polyscale analyses, and provides new insight in the theoretical and practical aspects of extracting the measure. © 2021 IEEE.

关键词： Signal analysis

来源：评论

学校读者我要写书评

暂无评论

Carrying Out CNN Channel Pruning in a White Box

引用

IEEE Transactions on Neural Networks and Learning systems 2023年第10期34卷 7946-7955页

作者： Zhang, Yuxin Lin, Mingbao Lin, Chia-Wen Chen, Jie Wu, Yongjian Tian, Yonghong Ji, Rongrong Xiamen University Media Analytics and Computing Laboratory Department of Artificial Intelligence School of Informatics Xiamen361005 China Youtu Laboratory Tencent Shanghai200233 China National Tsing Hua University Department of Electrical Engineering The Institute of Communications Engineering Hsinchu30013 Taiwan Peng Cheng Laboratory Shenzhen518000 China Peking University Department of Computer Science and Technology Beijing100871 China School of Informatics and the Institute of Artificial Intelligence Xiamen University Media Analytics and Computing Laboratory Department of Artificial Intelligence Xiamen361005 China

Channel pruning has been long studied to compress convolutional neural networks (CNNs), which significantly reduces the overall computation. Prior works implement channel pruning in an unexplainable manner, which tends to reduce the final classification errors while failing to consider the internal influence of each channel. In this article, we conduct channel pruning in a white box. Through deep visualization of feature maps activated by different channels, we observe that different channels have a varying contribution to different categories in image classification. Inspired by this, we choose to preserve channels contributing to most categories. Specifically, to model the contribution of each channel to differentiating categories, we develop a class-wise mask for each channel, implemented in a dynamic training manner with respect to the input image's category. On the basis of the learned class-wise mask, we perform a global voting mechanism to remove channels with less category discrimination. Lastly, a fine-tuning process is conducted to recover the performance of the pruned model. To our best knowledge, it is the first time that CNN interpretability theory is considered to guide channel pruning. Extensive experiments on representative image classification tasks demonstrate the superiority of our White-Box over many state-of-the-arts (SOTAs). For instance, on CIFAR-10, it reduces 65.23% floating point operations per seconds (FLOPs) with even 0.62% accuracy improvement for ResNet-110. On ILSVRC-2012, White-Box achieves a 45.6% FLOP reduction with only a small loss of 0.83% in the top-1 accuracy for ResNet-50. Code is available at https://***/zyxxμWhite-Box. © 2012 IEEE.

关键词： Visualization

来源：评论

学校读者我要写书评

暂无评论

Streamflow Prediction in the Mekong River Basin Using Deep Neural Networks

引用

IEEE Access 2023年 11卷 97930-97943页

作者： Nguyen, Thi-Thu-Ha Vu, Duc-Quang Mai, Son T. Dang, Thanh Duc National Taiwan University Department of Bioenvironmental Systems Engineering Taipei10617 Taiwan National Central University Department of Computer Science and Information System Taoyuan320317 Taiwan Thai Nguyen University of Education Thai Nguyen250000 Viet Nam Queen's University Belfast School of Electronics Electrical Engineering and Computer Science BelfastBT7 1NN United Kingdom Institute for Computational Science and Artificial Intelligence Van Lang University Laboratory of Environmental Sciences and Climate Change Ho Chi Minh City70000 Viet Nam Van Lang University Faculty of Environment Ho Chi Minh City70000 Viet Nam

In recent years, the Mekong River Basin (MRB), one of the largest river basins in Southeast Asia, has experienced severe impacts from extreme droughts and floods. Streamflow forecasting has become crucial for effective risk management strategies in the region. However, this task presents significant challenges due to rapid climate changes and the presence of numerous newly constructed upstream dams, which disrupt the natural flow. In this paper, we develop multiple deep learning models (incl. Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long short-term Memory (LSTM), and Transformer) to predict streamflow with different lead time forecasts based on observed meteorological variables and climatic indices (i.e., discharge, water level, precipitation, and temperature) from 1979 to 2019. The results indicate that LSTM obtains high performance for streamflow prediction in both dry and wet seasons while Transformer is not recommended for long-term prediction, especially in the dry season. The proposed deep learning models capture well the fluctuation of river flow in the MRB during the period of high-dam development, especially LSTM (NSE ≥ 0.8). The models' performances are enhanced with the adding of temperature for short-term prediction while precipitation was the most sensitive variable for long-term one. Such proposed models are essential for government agencies to plan mitigation and adaptation strategies at different periods, which can range from days to years. © 2013 IEEE.

关键词： Floods

来源：评论

学校读者我要写书评

暂无评论

Research on CMUTs Design with Different Cell Arrangements

引用

Recent Patents on engineering 2024年第1期18卷 110-118页

作者： Cao, Xuewen Yu, Yuanyu Liu, Xin Ma, Jingyi Zhang, Hua Wang, Lijia Intelligent Manufacturing Institute Hebei Vocational University of Industry and Technology Shi Jiazhuang050091 China School of Artificial Intelligence Neijiang Normal University Neijiang641100 China State Key Laboratory of Analog and Mixed-Signal VLSI University of Macau Macau 999078 China Department of Electrical and Computer Engineering Faculty of Science and Technology University of Macau Macau 999078 China

Background: Capacitive micromachined ultrasonic transducers (CMUTs) is a promising component of mechanical-electrical-acoustical conversion, which shows valuable applications in non-distructive testing and obstacle detection fields. In order to study the influence of dimension parameters on the directivity of CMUTs, the three-dimensional multi-physics coupling model is proposed for CMUTs in this paper. Methods: The model combines multiple physical fields through finite element method (FEM), which is more efficient and accurate. The influence of dimension parameters on the transducer is studied from two aspects: the pitch between adjacent cells and cell number, respectively. Results: When increasing the pitch, the directivity becomes better if the pitch is less than half of the wavelength, the grating lobe and side lobe are aroused. However, the bandwidth at-3dB is reduced by up to 71.1% at the same time. When increasing the cell number, the better the direcitvity, while the change of the cell number has little effect on the bandwidth of the array. However, 5.4% frequency shift was caused due to the effective mass change of the transducer. Conclusion: In this study, a 3D model based on FEM is proposed for the influence of dimension parameter on CMUTs. The simulation results indicate that the directivity can be enhanced by chang-ing the dimension. At the same time, it also introduces other performance issues, such as bandwidth decrease and frequency shift through this approach. Therefore, the simulation proposed the guidance of CMUTs performance optimization and patented technology application work in the future. © 2024, Bentham science Publishers. All rights reserved.

关键词： Finite element method

来源：评论

学校读者我要写书评

暂无评论

Multi-Stage Enhanced Diagnosis With Fault Candidate Reduction

引用

IEEE Transactions on computer-Aided Design of Integrated Circuits and systems 2025年

作者： Yun, Hyojoon Lim, Hyeonchan Lee, Hayoung Kang, Sungho Yonsei University Computer Systems Reliable SoC Laboratory Department of Electrical and Electronic Engineering Seoul03722 Korea Republic of Ajou University Department of Intelligence Semiconductor Engineering Suwon16499 Korea Republic of

Logic diagnosis is essential for improving reliability and yield. In conventional diagnosis methods, although various methods are proposed to enhance the accuracy and resolution of logic diagnosis, there are still diagnosis results where the reported locations of defects are incorrect. Particularly in logic circuits, which contain a large number of gates, multiple faults can occur, not just single faults. Since the number of possible cases for multiple faults is significantly greater compared to single faults, the diagnosis of multiple faults is complicated. To address this problem, a new diagnosis method that uses a multi-stage process with fault candidate reduction is proposed. In the proposed method, machine learning is used with fault candidate reduction, and post-processing is performed after the use of machine learning. This proposed method allows for the analysis of multiple faults using only the test responses for single faults, demonstrating that this method can maintain sufficient accuracy and resolution for unexpected faults. © 1982-2012 IEEE.

关键词： Logic gates

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：