检索结果-内蒙古大学图书馆

A Time-Series Neural Network for Pig Feeding Behavior Recognition and Dangerous Detection from Videos

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Zhang, Yan Yang, Xinze Zhou, Junyu Li, Jiapeng Zhang, Longxiang Ma, Qin College of Information and Electrical Engineering China Agricultural University Beijing100083 China Department of Computer Science and Technology Tsinghua University Beijing100084 China Department of Communication University of Colorado Denver Denver United States School of Computer Science and Engineering Beihang University Beijing China Fu Foundation School of Engineering and Applied Science Columbia University the City of New York New York City United States

With the development of modern animal husbandry, especially pig farming, large-scale, intensive, and automated farming has become the trend in the industry. The realization of accurate recognition and warning of dangerous actions in feeding scenarios for sows and piglets holds high research and practical value in this field. This paper addresses this issue by proposing a Transformer-based Neural Network (TNN) model. This model emphasizes the extraction of global features and handling of long-distance dependencies, significantly improving the accuracy of behavior recognition. Compared with traditional neural network models, TNN demonstrates superiority in dealing with animal behavior recognition in farming scenarios. Furthermore, an in-depth study of the attention mechanism in the TNN model was conducted in this paper. By visualizing the attention heat map of the TNN model, it was found that TNN could effectively focus on key areas in the image, thereby accurately identifying the behavior of piglets. Finally, this paper proposes a unique model lightweighting strategy that allows the TNN model to run efficiently on edge devices. In the experimental part, the performance of the TNN model on five behavior recognition tasks was evaluated first. The results showed that the TNN model achieved high scores in both Precision and Recall, far exceeding traditional neural network models. Then, by visualizing the attention heat map of the TNN model, the superiority of the TNN model in focusing on key image areas was further confirmed. In the end, the effect of the model lightweighting strategy was demonstrated, and even with a significant reduction in parameters and computational complexity, the performance of the TNN model remained excellent. This research not only promotes the development of animal behavior recognition technology but also provides new insights and tools for the precise management and efficient operation of the animal husbandry industry. © 2023, The Authors. All

关键词： Neural network models

Provably Convergent Plug-and-Play Quasi-Newton Methods

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Tan, Hong Ye Mukherjee, Subhadip Tang, Junqi Schönlieb, Carola-Bibiane Department of Applied Mathematics and Theoretical Physics University of Cambridge United Kingdom Department of Computer Science University of Bath United Kingdom School of Mathematics University of Birmingham United Kingdom Department of Electronics and Electrical Communication Engineering Indian Institute of Technology Kharagpur India

Plug-and-Play (PnP) methods are a class of efficient iterative methods that aim to combine data fidelity terms and deep denoisers using classical optimization algorithms, such as ISTA or ADMM, with applications in inverse problems and imaging. Provable PnP methods are a subclass of PnP methods with convergence guarantees, such as fixed point convergence or convergence to critical points of some energy function. Many existing provable PnP methods impose heavy restrictions on the denoiser or fidelity function, such as nonexpansiveness or strict convexity, respectively. In this work, we propose a novel algorithmic approach incorporating quasi-Newton steps into a provable PnP framework based on proximal denoisers, resulting in greatly accelerated convergence while retaining light assumptions on the denoiser. By characterizing the denoiser as the proximal operator of a weakly convex function, we show that the fixed points of the proposed quasi-Newton PnP algorithm are critical points of a weakly convex function. Numerical experiments on image deblurring and super-resolution demonstrate 2-8x faster convergence as compared to other provable PnP methods with similar reconstruction *** Codes 49M15, 49J52, 65K15 Copyright © 2023, The Authors. All rights reserved.

关键词： Image reconstruction

Fast Oexpected(N) Algorithm for Finding Exact Maximum Distance in E2 Instead of O(N2) or O(N lgN)

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Skala, Vaclav Department of Computer Science and Engineering Faculty of Applied Sciences University of West Bohemia PlzenCZ 306 14 Czech Republic

This paper describes novel and fast, simple and robust algorithm with O(N) expected complexity which enables to decrease run-time needed to find an exact maximum distance of two points in E2. The proposed algorithm has been evaluated experimentally on larger different datasets. The proposed algorithm gives a significant speed-up to applications, when medium and large data sets are processed. It is over 10 000 times faster than the standard algorithm for 106 points randomly distributed points in E2. Experiments proved the advantages of the proposed algorithm over the standard algorithm and convex hull diameters *** Codes 68xx, 68U05 © 2022, CC BY.

关键词： Pattern recognition

A Cascading Cooperative Multi-agent Framework for On-ramp Merging Control Integrating Large Language Models

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhang, Miao Fang, Zhenlong Wang, Tianyi Zhang, Qian Lu, Shuai Jiao, Junfeng Shi, Tianyu Shenzhen International Graduate School Tsinghua University Shenzhen518055 China Department of Computer Science & Engineering University of Minnesota Twin Cities MinneapolisMN55455 United States Department of Mechanical Engineering & Materials Science Yale University New HavenCT06511 United States Computer Vision BYD Company Limited Shenzhen China School of Architecture University of Texas at Austin AustinTX78712 United States Faculty of Applied Science & Engineering University of Toronto TorontoONM5S 1A1 Canada

Traditional Reinforcement Learning (RL) suffers from replicating human-like behaviors, generalizing effectively in multi-agent scenarios, and overcoming inherent interpretability issues. These tasks are compounded when deep environment understanding, agent coordination and dynamic optimization are required. While Large Language Model (LLM) enhanced methods have shown promise in generalization and interoperability, they often neglect necessary multi-agent coordination. Therefore, we introduce the Cascading Cooperative Multi-agent (CCMA) framework, integrating RL for individual interactions, a fine-tuned LLM for regional cooperation, a reward function for global optimization, and the Retrieval-augmented Generation mechanism to dynamically optimize decision-making across complex driving scenarios. Our experiments demonstrate that the CCMA outperforms existing RL methods, demonstrating significant improvements in both micro and macro-level performance in complex driving *** project page is https://***/***/ Copyright © 2025, The Authors. All rights reserved.

关键词： Multi agent systems

Locating saddle points using gradient extremals on manifolds adaptively revealed as point clouds

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Georgiou, A. Vandecasteele, H. Bello-Rivas, J.M. Kevrekidis, I. Department of Chemical and Biomolecular Engineering Johns Hopkins University BaltimoreMD21218 United States Department of Computer Science KU Leuven Leuven3001 Belgium Department of Applied Mathematics & Statistics Johns Hopkins University BaltimoreMD21218 United States

Steady states are invaluable in the study of dynamical systems. High-dimensional dynamical systems, due to a separation of time-scales, often evolve towards a lower dimensional manifold M. We introduce an approach to locate saddle points (and other fixed points) that utilizes gradient extremals on such a priori unknown (Riemannian) manifolds, defined by adaptively sampled point clouds, with local coordinates discovered on-the-fly through manifold learning. The technique, which efficiently biases the dynamical system along a curve (as opposed to exhaustively exploring the state space), requires knowledge of a single minimum and the ability to sample around an arbitrary point. We demonstrate the effectiveness of the technique on the Müller-Brown potential mapped onto an unknown surface (namely, a sphere). Previous work employed a similar algorithmic framework to find saddle points using Newton trajectories and gentlest ascent dynamics;we therefore also offer a brief comparison with these methods. Copyright © 2023, The Authors. All rights reserved.

关键词： Dynamical systems

Frequency Aware and Graph Fusion Network for Polyp Segmentation

学校读者我要写书评

暂无评论

Frequency Aware and Graph Fusion Network for Polyp Segmentat...

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Yan Li Zhuoran Zheng Wenqi Ren Yunfeng Nie Jingang Zhang Xiuyi Jia School of Future Technology University of Chinese Academy of Sciences Institute of Information Engineering Chinese Academy of Sciences School of Cyber Security University of Chinese Academy of Sciences School of Computer Science and Engineering Nanjing University of Science and Technology Department of Applied Physics and Photonics Vrije Universiteit Brussel and Flanders Make

Polyp segmentation plays a crucial role in the prevention of colon cancer. However, the diverse shapes of polyps and their similarity to normal areas in terms of color and texture make polyp segmentation a challenging task. Currently, most polyp segmentation methods solely focus on spatial domain features, ignoring the valuable features in the frequency domain. Consequently, many polyp segmentation algorithms struggle with the camouflage of polyps. To tackle this issue, we propose the Frequency Aware and Graph Fusion Network (FAGF-Net). Specifically, it begins with a Frequency-based Global Extraction Module (FGEM), which provides an initial estimation of the polyp regions to guide subsequent modules. Next, we design a Frequency-based Feature Attention Module (FFAM) that leverages amplitude and phase information to amplify appearance differences and enhance semantic representations. Moreover, we present a Graph-based Fusion Module (GFM), which infers the geometric characteristic of polyps through aggregating and interacting with enhanced features. Extensive experiments show that our method outperforms state-of-the-art methods with better quantitative and qualitative evaluations.

关键词：

Social Bots: Detection and Challenges

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Yang, Kai-Cheng Varol, Onur Nwala, Alexander C. Sayyadiharikandeh, Mohsen Ferrara, Emilio Flammini, Alessandro Menczer, Filippo Luddy School of Informatics Computing and Engineering Indiana University Bloomington United States Network Science Institute Northeastern University United States Faculty of Engineering and Natural Sciences Sabanci University Turkey Department of Applied Science William & Mary United States Thomas Lord Department of Computer Science University of Southern California United States

While social media are a key source of data for computational social science, their ease of manipulation by malicious actors threatens the integrity of online information exchanges and their analysis. In this Chapter, we focus on malicious social bots, a prominent vehicle for such manipulation. We start by discussing recent studies about the presence and actions of social bots in various online discussions to show their real-world implications and the need for detection methods. Then we discuss the challenges of bot detection methods and use Botometer, a publicly available bot detection tool, as a case study to describe recent developments in this area. We close with a practical guide on how to handle social bots in social media research. © 2023, CC BY.

关键词： Botnet

Reconfigurable inverse-designed phase-change photonics

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wu, Changming Jiao, Ziyu Deng, Haoqin Huang, Yi-Siou Yu, Heshan Takeuchi, Ichiro Ríos Ocampo, Carlos A. Li, Mo Department of Electrical and Computer Engineering University of Washington SeattleWA98195 United States Department of Materials Science and Engineering University of Maryland College ParkMD20742 United States Institute for Research in Electronics and Applied Physics University of Maryland College ParkMD20742 United States Department of Physics University of Washington SeattleWA98195 United States

Chalcogenide phase-change materials (PCMs) offer a promising approach to programmable photonics thanks to their nonvolatile, reversible phase transitions and high refractive index contrast. However, conventional designs are limited by global phase control over entire PCM thin films between fully amorphous and fully crystalline states, which restricts device functionality and confines design flexibility and programmability. In this work, we present a novel approach that leverages pixel-level control of PCM in inverse-designed photonic devices, enabling highly reconfigurable, multi-functional operations. We integrate low-loss Sb2Se3 onto a multi-mode interferometer (MMI) and achieve precise, localized phase manipulation through direct laser writing. This technique allows for flexible programming of the photonic device by adjusting the PCM phase pattern rather than relying on global phase states, thereby enhancing device adaptability. As a proof of concept, we programmed the device as a wavelength-division multiplexer and subsequently reconfigured it into a mode-division multiplexer. Our results underscore the potential of combining inverse design with pixel-wise tuning for next-generation programmable phase-change photonic systems. © 2024, CC BY.

关键词： Amorphous films

SecFedMDM-1: A Federated Learning-Based Malware Detection Model for Interconnected Cloud Infrastructures

学校读者我要写书评

暂无评论

IEEE Access 2025年 13卷 101246-101261页

作者： Daniella Mughole Kalimumbalo Joke A. Badejo Kennedy O. Okokpujie Emmanuel Adetiba Molo Mbasa Joaquim Nzanzu Vingi Patrick Gabriel Oyeyemi Claude Takenga Department of Electrical and Information Engineering College of Engineering Covenant University Ota Nigeria Covenant Applied Informatics and Communication African Centre of Excellence Covenant University Ota Nigeria Département de Génie Electrique et Informatique Faculté des Sciences et Technologie Appliquées Université Libre des Pays des Grands Lacs Goma Democratic Republic of the Congo HRA Industrial Engineering Department Durban University of Technology Durban South Africa Department of Computer and Information Science College of Science and Technology Covenant University Ota Nigeria Entreprise NTIC Infokom GmbH Neubrandenburg Germany

Enforcing security and reliability in the cloud is a challenging but vital task because of the multitude of heterogeneous applications utilising the same infrastructure. Integrating a security analysis system to identify and mitigate potential threats, such as malicious software (Malware), is crucial for the effective functioning of cloud infrastructures. Over the past few decades, new malware analysis and detection approaches have emerged because of various malware strategies that evade host- and network-based security measures. This study established a federated learning-based malware detection model for interconnected cloud infrastructures. This method protects users’ privacy as numerous devices can collaborate to build machine learning models without sharing data. Three distinct deep-learning algorithms were chosen for the models’ training, validation, and testing phases. With the training of eight clients and twenty-five federation rounds, the FeedForward Neural Networks (FFNN) model performed best. It had accuracy, an F1-score, and a precision of 84%, while the Multi-Layer Perceptron (MLP) model performed with 83% accuracy, 83% F1-score, and 83% precision, and the Long Short-Term Memory (LSTM) model performed with 80% accuracy, 80% F1-score, and 80% precision.

关键词： Malware Cloud computing computer security Companies Data models Servers Federated learning Training Data security Data privacy