检索结果-内蒙古大学图书馆

Control-based Graph Embeddings with Data Augmentation for Contrastive Learning

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ahmad, Obaid Ullah Said, Anwar Shabbir, Mudassir Abbas, Waseem Koutsoukos, Xenofon Electrical Engineering Department University of Texas at Dallas RichardsonTX United States Computer Science Department Vanderbilt University NashvilleTN United States Systems Engineering Department University of Texas at Dallas RichardsonTX United States

In this paper, we study the problem of unsupervised graph representation learning by harnessing the control properties of dynamical networks defined on graphs. Our approach introduces a novel framework for contrastive learning, a widely prevalent technique for unsupervised representation learning. A crucial step in contrastive learning is the creation of 'augmented' graphs from the input graphs. Though different from the original graphs, these augmented graphs retain the original graph's structural characteristics. Here, we propose a unique method for generating these augmented graphs by leveraging the control properties of networks. The core concept revolves around perturbing the original graph to create a new one while preserving the controllability properties specific to networks and graphs. Compared to the existing methods, we demonstrate that this innovative approach enhances the effectiveness of contrastive learning frameworks, leading to superior results regarding the accuracy of the classification tasks. The key innovation lies in our ability to decode the network structure using these control properties, opening new avenues for unsupervised graph representation learning. © 2024, CC BY.

关键词： Graphic methods

CapsEnhancer: An Effective Computational Framework for Identifying Enhancers Based on Chaos Game Representation and Capsule Network

学校读者我要写书评

暂无评论

Journal of Chemical Information and Modeling 2024年第14期64卷 5725-5736页

作者： Yao, Lantian Xie, Peilin Guan, Jiahui Chung, Chia-Ru Huang, Yixian Pang, Yuxuan Wu, Huacong Chiang, Ying-Chih Lee, Tzong-Yi Kobilka Institute of Innovative Drug Discovery School of Medicine The Chinese University of HongKong Shenzhen518172 China School of Science and Engineering The Chinese University of HongKong Shenzhen518172 China School of Medicine The Chinese University of HongKong Shenzhen518172 China Department of Computer Science and Information Engineering National Central University Taoyuan320317 Taiwan Division of Health Medical Intelligence Human Genome Center The University of Tokyo Tokyo108-8639 Japan Institute of Bioinformatics and Systems Biology National Yang Ming Chiao Tung University Hsinchu300093 Taiwan National Yang Ming Chiao Tung University Hsinchu300093 Taiwan

Enhancers are a class of noncoding DNA, serving as crucial regulatory elements in governing gene expression by binding to transcription factors. The identification of enhancers holds paramount importance in the field of biology. However, traditional experimental methods for enhancer identification demand substantial human and material resources. Consequently, there is a growing interest in employing computational methods for enhancer prediction. In this study, we propose a two-stage framework based on deep learning, termed CapsEnhancer, for the identification of enhancers and their strengths. CapsEnhancer utilizes chaos game representation to encode DNA sequences into unique images and employs a capsule network to extract local and global features from sequence "images". Experimental results demonstrate that CapsEnhancer achieves state-of-the-art performance in both stages. In the first and second stages, the accuracy surpasses the previous best methods by 8 and 3.5%, reaching accuracies of 94.5 and 95%, respectively. Notably, this study represents the pioneering application of computer vision methods to enhancer identification tasks. Our work not only contributes novel insights to enhancer identification but also provides a fresh perspective for other biological sequence analysis tasks. © 2024 The Authors. Published by American Chemical Society.

关键词： DNA sequences

An Intelligence Defense System with SNORT Rules

学校读者我要写书评

暂无评论

An Intelligence Defense System with SNORT Rules

International Conference on Advanced Communication Technology (ICACT)

作者： Yi-Cheng Lai Chiao-Lin Yu Man-Ling Liao Yu-Shan Lin Yao-Chung Chang Jiann-Liang Chen Department of Electrical Engineering National Taiwan University of Science and Technology Taipei Taiwan Department of Information Science and Management Systems National Taitung University Taitung Taiwan Department of Computer Science and Information Engineering National Taitung University Taitung Taiwan

Misconfiguration of firewall rules has always been considered a serious issue. The handwritten rule is messy and buggy under the increasingly complex firewall architecture. To avoid being attacked behind an insecure firewall. This study defines an intelligence defense system. Combined with data analysis, feature extraction, optimization, and firewall technology. Its main purpose is to replace handwritten firewall rules and provide immediate and reliable protection against diversified attacks. In the verification, 68,936,206 packets collected by Cowrie honeypot were used as the test data. The accuracy rate of classifying different attack behaviors reached 99.5%, and the packet coverage of Snort rules also achieved 98%. This thesis proposes a system that can effectively identify and defend from diverse attacks.

关键词： Data analysis Firewalls (computing) Feature extraction Real-time systems Communications technology Classification algorithms Behavioral sciences

Automated Bi-Fold Weighted Ensemble Algorithms and its Application to Brain Tumor Detection and Classification

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Huang, PoTsang B. Rizwan, Muhammad Ali, Mehboob Department of Electrical Engineering and Computer Science National Taiwan University Taipei Taiwan Department of Industrial and Systems Engineering Chung Yuan Christian University Zhongli Taiwan Department of Applied Mathematics Chung Yuan Christian University Zhongli Taiwan

The uncontrolled and unstructured growth of brain cells is known as brain tumor, which has one of the highest mortality rates among diseases from all types of cancers. Due to limited diagnostic and treatment capabilities, they pose significant challenges, especially in third-world countries. Early diagnosis plays a vital role in effectively managing brain tumors and reducing mortality rates. However, the availability of diagnostic methods is hindered by various limitations, including high costs and lengthy result acquisition times, impeding early detection of the disease. In this study, we present two cutting-edge bi-fold weighted voting ensemble models that aim to boost the effectiveness of weighted ensemble methods. These two proposed methods combine the classification outcomes from multiple classifiers and determine the optimal result by selecting the one with the highest probability in the first approach, and the highest weighted prediction in the second technique. These approaches significantly improve the overall performance of weighted ensemble techniques. In the first proposed method, we improve the soft voting technique (SVT) by introducing a novel unsupervised weight calculating schema (UWCS) to enhance its weight assigning capability, known as the extended soft voting technique (ESVT). Secondly, we propose a novel weighted method (NWM) by using the proposed UWCS. Both of our approaches incorporate three distinct models: a custom-built CNN, VGG-16, and InceptionResNetV2 which has been trained on publicly available datasets. The effectiveness of our proposed systems is evaluated through blind testing, where exceptional results are achieved. We then establish a comparative analysis of the performance of our proposed methods with that of SVT to show their superiority and effectiveness. © 2024, CC BY.

关键词： Diseases

Temperature Dependence of Hot-Electron Graphene Fet Bolometric Detectors Response to Modulated Terahertz Radiation

学校读者我要写书评

暂无评论

International Journal of High Speed Electronics and systems 2024年第4期33卷

作者： Ryzhii, Maxim Ryzhii, Victor Tang, Chao Otsuji, Taiichi Mitin, Vladimir Shur, Michael S. Department Of Computer Science And Engineering University Of Aizu Aizu-Wakamatsu Fukushima965-8580 Japan Research Institute For Electrical Communication Tohoku University Miyagi Sendai980-8577 Japan Department Of Electrical Engineering University At Buffalo Suny BuffaloNY14260 United States Department Of Electrical Computer And Systems Engineering Rensselaer Polytechnic Institute TroyNY12180 United States

In this paper, we analyze the modulation characteristics and the ultimate modulation frequency of the terahertz (THz) hot-electron FET bolometers with the graphene channels (GCs), metal gate (MG), and gate barrier layers (BLs) in a wide temperature range. Our results predict that the responsivity of GC-FET bolometers decreases with decreasing operating temperature. This is attributed to a dramatic drop in the thermionic GC-MG current (characterized by the relatively large activation energy) and its modulated component when the temperature lowers. A further decrease in the temperature results in a fairly strong responsivity roll-off. In contrast, the responsivity of the GC-FET detectors with the temperature-adjusted load resistance rises with decreasing temperature because the temperature lowering leads to an increase in the electron energy relaxation time and promotes more effective heating by the impinging THz radiation. In this case, the weakening of the thermionic current is compensated by the commensurate increase in the load resistance. © 2024 World Scientific Publishing Company.

关键词： Graphene

Discriminative Finetuning of Generative Large Language Models without Reward Models and Preference Data

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Guo, Siqi Hong, Ilgee Balmaseda, Vicente Zhao, Tuo Yang, Tianbao Department of Computer Science and Engineering Texas A&M University College Station United States H. Milton Stewart School of Industrial and Systems Engineering Georgia Institute of Technology Atlanta United States

Supervised fine-tuning (SFT) followed by preference optimization (PO) denoted by SFT→PO has become the standard for improving pretrained large language models (LLMs), with PO demonstrating significant performance gains. However, PO methods rely on either human-labeled preference data or a strong reward model to generate preference data. Can we fine-tune LLMs without preference data or reward models while achieving competitive performance to SFT→PO? We address this question by introducing Discriminative Fine-Tuning (DFT), a novel approach that eliminates the need for preference data. Unlike SFT, which employs a generative approach and overlooks negative data, DFT adopts a discriminative paradigm that that increases the probability of positive answers while suppressing potentially negative ones, shifting from token prediction to data prediction. Our contributions include: (i) a discriminative probabilistic framework for fine-tuning LLMs by explicitly modeling the discriminative likelihood of an answer among all possible outputs given an input;(ii) efficient algorithms to optimize this discriminative likelihood;and (iii) extensive experiments demonstrating DFT’s effectiveness, achieving performance better than SFT and comparable to if not better than SFT→PO. The code can be found at https://***/PenGuln/DFT. © 2025, CC BY.

关键词： Digital elevation model

Data-driven Optimal Cooperative Output Regulation of Unknown Linear Discrete-time Multi-agent systems

学校读者我要写书评

暂无评论

Data-driven Optimal Cooperative Output Regulation of Unknown...

第35届中国控制与决策会议

作者： Dong Liang Yi Dong Chaoli Wang Engang Tian Department of Control Science and Engineering School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology College of Electronic and Information Engineering Shanghai Research Institute for Intelligent Autonomous Systems Tongji University

The cooperative output regulation problem has been extensively studied on the basis of the distributed observer ***,the majority of the existing research assumes that the dynamics is known *** remove this condition,the cooperative output regulation problem is further solved via the data-driven framework where the dynamics of the plant is ***,a data-driven distributed observer is established to estimate the state of the leader with unknown dynamics subject to external ***,the unknown regulator equations are solved using the iterative recurrent neural network ***,the state-based data-driven distributed control law is synthesized to solve the *** optimal gains are derived by solving convex optimization problems using input and state ***,a numerical example is presented to verify the feasibility of the proposed framework.

关键词： Linear systems multi-agent systems cooperative output regulation data-driven control optimal control

Multi-Unit Floor Plan Recognition and Reconstruction Using Improved Semantic Segmentation of Raster-Wise Floor Plans

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Kratochvila, Lukas de Jong, Gijs Arkesteijn, Monique Bilík, Šimon Zemčík, Tomáš Horak, Karel Rellermeyer, Jan S. Department of Control and Instrumentation Faculty of Electrical Engineering and Communication Brno University of Technology Brno Czech Republic Department of Software Technology Faculty of Electrical Engineering Mathematics and Computer Science TU Delft Delft Netherlands Department of Management in the Built Environment Faculty of Architecture and the Built Environment TU Delft Delft Netherlands Computer Vision and Pattern Recognition Laboratory Department of Computational Engineering Lappeenranta-Lahti University of Technology LUT Lappeenranta Finland Dependable and Scalable Software Systems Institute of Systems Engineering Faculty of Electrical Engineering and Computer Science Leibniz University Hannover Hannover Germany

Digital twins have a major potential to form a significant part of urban management in emergency planning, as they allow more efficient designing of the escape routes, better orientation in exceptional situations, and faster rescue intervention. Nevertheless, creating the twins still remains a largely manual effort, due to a lack of 3D-representations, which are available only in limited amounts for some new buildings. Thus, in this paper we aim to synthesize 3D information from commonly available 2D architectural floor plans. We propose two novel pixel-wise segmentation methods based on the MDA-Unet and MACU-Net architectures with improved skip connections, an attention mechanism, and a training objective together with a reconstruction part of the pipeline, which vectorizes the segmented plans to create a 3D model. The proposed methods are compared with two other state-of-the-art techniques and several benchmark datasets. On the commonly used CubiCasa benchmark dataset, our methods have achieved the mean F1 score of 0.86 over five examined classes, outperforming the other pixel-wise approaches tested. We have also made our code publicly available to support research in the field. © 2024, CC BY.

关键词： Semantic Segmentation

A Deep Learning Framework for Healthy LifeStyle Monitoring and Outdoor Localization

学校读者我要写书评

暂无评论

IEEE Access 2025年

作者： Rafiq, Mehrab Alshammari, Naif S. Alhasson, Haifa F. AlHammadi, Dina Abdulaziz Alshehri, Mohammed Jalal, Ahmad Liu, Hui Air University Department of Computing and AI E-9 Islamabad44000 Pakistan Majmaah University College of Computer and Information Sciences Department of Computer Sciences Majmaah11952 Saudi Arabia Qassim University College of Computer Department of Information Technology Buraydah52571 Saudi Arabia Princess Nourah bint Abdulrahman University Department of Information Systems College of Computer and Information Sciences P.O.Box 84428 Riyadh11671 Saudi Arabia King Khalid University Department of Computer Science Abha61421 Saudi Arabia Korea University College of Informatics Department of Computer Science and Engineering Seoul02841 Korea Republic of University of Bremen Cognitive Systems Lab Bremen28359 Germany

The green research field of ubiquitous computing has been able to draw and hold academics’ interest for a while. Recognition and localization of human locomotion have also been widely developed as ubiquitous computing applications. Personal safety, behavior analysis, entertainment, and healthcare monitoring all utilize these apps.A key component of several fields, such as robots, sports, healthcare, and security, is human locomotion recognition (HLR). Researchers and engineers have been trying to use the increasing popularity of wearable technology, especially environmental sensors and Inertial Measurement Units (IMUs), to identify and categorize human locomotion activities in an accurate and efficient *** capabilities of smartphones and wearable technology have increased due to advancements in sensing technology. Inertial sensors like gyroscopes and accelerometers are now frequently seen in smartphones. These sensors can now be utilized for a wide range of purposes, while their original purpose was to improve gadget features. Using smartphone IMU, ambient, GPS, and audio sensor data from two publicly available benchmark datasets—the Extrasensory dataset and the Domino dataset—this study proposes a sophisticated approach for human locomotion and localization detection. In the preprocessing stage, a Chebyshev Type 2 filter was used for windowing and segmentation, while a Hamming window was applied. Feature extraction was divided into two parts: for actions, the extracted features included Fast Fourier Transform (FFT), State Space Correlation Entropy (SSCE), Maximum Lyapunov Exponent (MLE), and Auto Regression;for localization-based features, Recursive Feature Elimination (RFE), step count, heading angle, and step length were employed. Kernel Fisher Discriminant Analysis was applied for feature optimization, and a deep neural network was utilized for feature classification. The proposed system achieved an overall classification accuracy of 88.4% on the Extrasen

关键词： Ambient intelligence