检索结果-内蒙古大学图书馆

26th International Conference on Information Fusion, FUSION 2023

作者： Zhang, Tianhao Wang, Shenglin Bouaynaya, Nidhal Calinescu, Radu Mihaylova, Lyudmila The University of Sheffield Automatic Control and Systems Engineering United Kingdom Rowan University Department of Electrical and Computer Engineering United States University of York Department of Computer Science United Kingdom

ISBN: (纸本)9798890344854

The superior performance of object detectors is often established under the condition that the test samples are in the same distribution as the training data. However, in many practical applications, out-of-distribution (OOD) instances are inevitable and usually lead to uncertainty in the results. In this paper, we propose a novel, intuitive, and scalable probabilistic object detection method for OOD detection. Unlike other uncertainty-modeling methods that either require huge computational costs to infer the weight distributions or rely on model training through synthetic outlier data, our method is able to distinguish between in-distribution (ID) data and OOD data via weight parameter sampling from proposed Gaussian distributions based on pre-trained networks. We demonstrate that our Bayesian object detector can achieve satisfactory OOD identification performance by reducing the FPR95 score by up to 8.19% and increasing the AUROC score by up to 13.94% when trained on BDD100k and VOC datasets as the ID datasets and evaluated on COCO2017 dataset as the OOD dataset. © 2023 International Society of Information Fusion.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Supervisors for Gas Compressor Stations with Compression and Valve Faults 8

Supervisors for Gas Compressor Stations with Compression and...

引用

8th South-East Europe Design Automation, computer Engineering, computer Networks and Social Media Conference, SEEDA-CECNSM 2023

作者： Fragkoulis, Dimitrios G. Koumboulis, Fotis N. Kouvakas, Nikolas D. Menexis, Antonis N. School of Science National and Kapodistrian University of Athens Robotics Automatic Control and Cyber-Physical Systems Laboratory Department of Digital Industry Technologies 34400 Greece

ISBN: (纸本)9798350383676

A Discrete Event System (DES) based approach is developed for the derivation of the model and safe performance of a gas compressor station (GCS), in the presence of all possible actuator faults. The safe performance goal is surge avoidance, despite the presence of actuator faults, and is modelled as four rules, being translated to four regular languages. The respective supervisor automata are developed to realize the four regular languages. The physical realizability and nonblocking of the synchronous product of the supervisor automata and the automaton of the system is proved. Implementation of the supervisors in the Structured Text language for PLCs is provided. © 2023 IEEE.

关键词： Discrete event simulation

来源：评论

学校读者我要写书评

暂无评论

Binary Stochastic Flip Optimization for Training Binary Neural Networks

Binary Stochastic Flip Optimization for Training Binary Neur...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Shibuya, Tatsukichi Inoue, Nakamasa Kawakami, Rei Sato, Ikuro Department of Computer Science Institute of Science Tokyo Tokyo Japan Department of Systems and Control Engineering Institute of Science Tokyo Tokyo Japan Denso IT Laboratory Tokyo Japan

ISBN: (纸本)9798350368741

For deploying deep neural networks on edge devices with limited resources, binary neural networks (BNNs) have attracted significant attention, due to their computational and memory efficiency. However, once a neural network is binarized, finetuning it on edge devices becomes challenging because most conventional training algorithms for BNNs are designed for use on centralized servers and require storing real-valued parameters during training. To address this limitation, this paper introduces binary stochastic flip optimization (BinSFO), a novel training algorithm for BNNs. BinSFO employs a parameter update rule based on Boolean operations, eliminating the need to store real-valued parameters and thereby reducing memory requirements and computational overhead. In experiments, we demonstrated the effectiveness and memory efficiency of BinSFO in finetuning scenarios on six image classification datasets. BinSFO performed comparably to conventional training algorithms with a 70.7% smaller memory requirement. Code is released at https://***/TatsukichiShibuya/ICASSP2025_BinSFO. © 2025 IEEE.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Optimizing Artificial Neural Networks Trough Weight Adjustments

引用

Procedia computer science 2024年 246卷 2158-2165页

作者： Syed Muhammad Abrar Akber Agnieszka Szczesna Sadia Nishat Kazmi Department of Computer Graphics Vision and Digital Systems Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice 44-100 Poland Faculty of Automatic Control Electronics and Computer Science Silesian University of Technology Gliwice 44-100 Poland

Longer training times pose a significant challenge in Artificial neural networks (ANNs) as it may leads to increasing the computational costs and decreasing the effectiveness of the model. Therefore, it is imperative to reduce training times in ANNs to enhance the computational efficiency. The initialization of the weights between the layers in ANN plays a vital role in reducing training times. Appropriate weight initialization can help the network converge faster during the training by providing an optimum starting point for the network. Therefore, weight initialization techniques are essential for efficient training of ANNs. This paper revisits and implements different popular weight initialization techniques in ANNs and analyzes their impact on training time. Specifically, this paper implements Gaussian-based, Kaming-based, and Xavier-based weight initiation atop a popular DNN-based network. The experiments are conducted by employing a well-known dataset. The results show that the scenario when no weight initiation is applied consumed the highest training time, whereas different weight initiation techniques contribute in reducing the training times for the network.

关键词： Artificial neural network Deep neural network weight initiation

来源：评论

学校读者我要写书评

暂无评论

Physics-Guided State-Space Model Augmentation Using Weighted Regularized Neural Networks ⁎

引用

IFAC-PapersOnLine 2024年第15期58卷 295-300页

作者： Yuhan Liu Roland Tóth Maarten Schoukens Control Systems Group Eindhoven University of Technology Eindhoven the Netherlands Systems and Control Laboratory Institute for Computer Science and Control Budapest Hungary

Physics-guided neural networks (PGNN) is an effective tool that combines the benefits of data-driven modeling with the interpretability and generalization of underlying physical information. However, for a classical PGNN, the penalization of the physics-guided part is at the output level, which leads to a conservative result as systems with highly similar state-transition functions, i.e. only slight differences in parameters, can have significantly different time-series outputs. Furthermore, the classical PGNN cost function regularizes the model estimate over the entire state space with a constant trade-of hyperparameter. In this paper, we introduce a novel model augmentation strategy for nonlinear state-space model identification based on PGNN, using a weighted function regularization (W-PGNN). The proposed approach can efficiently augment the prior physics-based state-space models based on measurement data. A new weighted regularization term is added to the cost function to penalize the difference between the state and output function of the baseline physics-based and final identified model. This ensures the estimated model follows the baseline physics model functions in regions where the data has low information content, while placing greater trust in the data when a high informativity is present. The effectiveness of the proposed strategy over the current PGNN method is demonstrated on a benchmark example.

关键词： System Identification Physics-Guided Neural Networks State Space

来源：评论

学校读者我要写书评

暂无评论

Scientific Research Transformation Needed for Comprehensive Artificial Intelligence Implementation 17

Scientific Research Transformation Needed for Comprehensive ...

引用

17th International Conference on Management of Large-Scale System Development, MLSD 2024

作者： Medennikov, Viktor Bogatyreva, Larisa Federal Research Center 'Computer Science and Control' of the Russian Academy of Sciences Information and Computing Systems Department Moscow Russia V.A. Trapeznikov Institute of Control Sciences of the Russian Academy of Sciences Laboratory of Modular Data Processing and Control Systems Moscow Russia

ISBN: (纸本)9798350375718

The paper deals with complementary changes in scientific research and real economy (exemplified by the farming industry) taking place when using the holistic approach to the industry's digital transformation resulting in the appearance of precise agrarian technologies that require a comprehensive, system-based combination of research and production resources that are able to ensure increased productivity in the industry. We demonstrate that among these new technologies, the artificial intelligence methods have a special significance. They have to, however, go through integration transformations to become standards for the proposed common digital farming management platform. © 2024 IEEE.

关键词： Agricultural economics

来源：评论

学校读者我要写书评

暂无评论

Robust control and data reconstruction for nonlinear epidemiological models using feedback linearization and state estimation

引用

Mathematical Biosciences and Engineering 2025年第1期22卷 109-137页

作者： Csutak, Balázs Szederkényi, Gábor Faculty of Information Technology and Bionics Pázmány Péter Catholic University Práter u. 50/A BudapestH-1083 Hungary Systems and Control Laboratory HUN-REN Institute for Computer Science and Control Kende u. 13-17 BudapestH-1111 Hungary

It has been clearly demonstrated over the past years that control theory can provide an efficient framework for the solution of several complex tasks in epidemiology. In this paper, we present a computational approach for the state estimation based reference tracking control and historical data reconstruction using nonlinear compartmental epidemic models. The control model is given in nonlinear input-affine form, where the manipulable input is the disease transmission rate influenced by possible measures and restrictions, while the observed or computed output is the number of infected people. The control design is built around a simple SEIR model and relies on a feedback linearization technique. We examine and compare different control setups distinguished by the availability of state information, complementing the directly measurable data with an extended Kalman filter used for state estimation. To illustrate the capabilities and robustness of the proposed method, we carry out multiple case studies for output tracking and data reconstruction on Swedish and Hungarian data, all in the presence of serious model and parameter mismatch. Computation results show that a well-designed feedback, even in the presence of significant observation uncertainties, can sufficiently reduce the effect of modeling errors. © 2025 the Authors,

关键词： Feedback linearization

来源：评论

学校读者我要写书评

暂无评论

Tracking and Mapping Strategy for Indoor UAV Based on Entropy Theory:An ORB-SLAM3 Extension 41

Tracking and Mapping Strategy for Indoor UAV Based on Entrop...

引用

第41届中国控制会议

作者： Jian Xu Haoyu Qi Min Xu Yunge Zang Zhen Li Xi Zhang Xiangdong Liu School of Automation and Key Laboratory for Intelligent Control & Decision on Complex Systems Beijing Institute of Technology Beijing Aerospace Automatic Control Institute National Key Laboratory of Science and Technology on Aerospace Intelligence Control

ISBN: (数字)9789887581536

ISBN: (纸本)9781665482561

For indoor application,which is an entirely GPS-denied environment,visual simultaneous localization and mapping(SLAM) facilitates the real autonomy of unmanned aerial vehicle(UAV) but raises the challenging requirement on accurate localization with low computational *** address this difficulty,a stereo-camera based SLAM system is proposed by applying Entropy theory to *** an extension to ORB-SLAM3,the additional entropy decision module and map processor are specifically *** decision module can improve computing efficiency by deciding whether keyframes or extra optimization should be ***,the map processor is targeted at loading and maintaining the prior map whenever *** results in indoor laboratory environment show that the developed system can achieve the superior localization accuracy in more efficient computation manner with smaller size of mapping compared with ***,the map can be effectively expanded and corrected even when prior information is invalid,greatly increasing the robustness of SLAM system.

关键词： UAV stereo-camera ORB-SLAM3 entropy theory prior map

来源：评论

学校读者我要写书评

暂无评论

A Non-Parametric Identification Scheme of SISO-LTI systems Using Blaschke-Products∗ 32

A Non-Parametric Identification Scheme of SISO-LTI Systems U...

引用

32nd Mediterranean Conference on control and Automation, MED 2024

作者： Ungvari, Gergo Soumelidis, Alexandros Dozsa, Tamas Eötvös Loránd University Faculty of Informatics Department of Numerical Analysis Budapest Hungary HUN-REN Institute for Computer Science and Control Systems and Control Laboratory Budapest Hungary

ISBN: (纸本)9798350395440

In this manuscript, a novel algorithm is presented for the identification of single input single output linear time invariant (SISO-LTI) systems. The proposed method is able to find poles of the transfer function describing the system without any assumptions on the system order. The results presented in this paper generalize earlier identification schemes based on expansions of the transfer function using generalized rational orthogonal basis functions (GOBFs). These previous methods depended on a convergent sequence constructed from the coefficients of the orthogonal GOBF expansion. The first contribution of this study is showing that a wider family of functions may be used to construct similar pole finding algorithms. The aforementioned previous approaches can be interpreted as special cases of the results presented here. In addition, using this generalization a new pole identification scheme is proposed using finite Blaschke-products. Through experiments, the practical utility of this scheme is verified when compared to earlier variations of the pole finding scheme. © 2024 IEEE.

关键词： Poles

来源：评论

学校读者我要写书评

暂无评论

Seek-and-Take Games of Heterogeneous Agent Teams With Large Language Model

Seek-and-Take Games of Heterogeneous Agent Teams With Large ...

引用

2024 China Automation Congress, CAC 2024

作者： Li, Haolin Yi, Peng Wei, Dixiao Bai, Wenyan The Department of Control Science and Engineering Tongji University Shanghai201804 China Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai201203 China National Key Laboratory of Autonomous Intelligent Unmanned Systems Frontiers Science Center for Intelligent Autonomous Systems Ministry of Education China Beijing Aerospace Automatic Control Institute Beijing100854 China

ISBN: (纸本)9798350368604

The problem of confrontation games between multi-agent teams has attracted considerable interest, and the question of how to ensure effective coordination of heterogeneous agents in dynamic and adversarial environments remains a significant challenge. In this paper, we present a decision-making methodology for a confrontation game between two teams, which is based on large language models (LLMs). The game is designed as a competitive Seek-and-Take task comprising three distinct robot types: reconnaissance, grabbing, and blocking robots. The LLMs enhance the robot team’s decision-making efficiency, delivering high-level commands that significantly improve coordination and adversarial performance. The game is initially formalized as a partially observable Markov decision process. Subsequently, a systematic construction of the LLM agents is provided, enabling real-time decisions based on the state information of a heterogeneous robot team (HRT). Detailed prompt design and the "Zero-shot-CoT" procedure are provided, and a cue word iteration method with reflections is proposed as a means of enhancing the decision-making efficiency of the LLMs and generating optimal collaboration strategies. Finally, the efficacy of the methodology is validated through experimentation with multiple LLMs in a multi-robot simulation environment. The results demonstrate that LLMs improve the efficiency and adaptability of decision-making in dynamic environments by serving as a core decision module, generating optimal strategies through real-time analysis. © 2024 IEEE.

关键词： Efficiency

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：