检索结果-内蒙古大学图书馆

2nd International Conference on Network Simulation and Evaluation, NSE 2023

作者： Bai, Youming Li, Shudong Han, Weihong Wu, Xiaobo Cyberspace Institute of Advanced Technology Guangzhou University Guangzhou510006 China Department of New Networks Peng Cheng Laboratory Shenzhen518000 China School of Computer Science and Cyber Engineering Guangzhou University Guangzhou510006 China

ISBN: (纸本)9789819745210

Malware, as one of the main threats to computer systems and network security, is constantly evolving and growing, causing serious impacts on individual users, business organizations, and national security. This review summarizes and sorts out the current research status in the field of malware detection, focusing on the file structure, common types of malware, malware camouflage techniques, common analysis methods and techniques for malware detection, including feature based static analysis, dynamic analysis, machine learning methods, etc. Furthermore, this article discusses the methods and applications of malware detection, and analyzes some methods and ideas for detecting malware based on different features. Through the review of this article, readers can comprehensively understand the current status and development trends of malware detection, providing reference and guidance for improving network security protection capabilities. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Once Read is Enough: Domain-specific Pretraining-free Language Models with Cluster-guided Sparse Experts for Long-tail Domain Knowledge 38

Once Read is Enough: Domain-specific Pretraining-free Langua...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Dong, Fang Chen, Mengyi Zhou, Jixian Shi, Yubin Chen, Yixuan Dong, Mingzhi Wang, Yujiang Li, Dongsheng Yang, Xiaochen Zhu, Rui Dick, Robert Lv, Qin Yang, Fan Lu, Tun Gu, Ning Shang, Li China and Shanghai Key Laboratory of Data Science School of Computer Science Fudan University China Oxford Suzhou Centre for Advanced Research Suzhou China University of Bath United Kingdom Microsoft Research Asia Shanghai China School of Mathematics & Statistics University of Glasgow United Kingdom Bayes Business School City St George's University of London United Kingdom Department of Electrical Engineering and Computer Science University of Michigan United States Department of Computer Science University of Colorado Boulder United States School of Microelectronics Fudan University China

Language models (LMs) only pretrained on a general and massive corpus usually cannot attain satisfying performance on domain-specific downstream tasks, and hence, applying domain-specific pretraining to LMs is a common and indispensable practice. However, domain-specific pretraining can be costly and time-consuming, hindering LMs' deployment in real-world applications. In this work, we consider the incapability to memorize domain-specific knowledge embedded in the general corpus with rare occurrences and "long-tail" distributions as the leading cause for pretrained LMs' inferior downstream performance. Analysis of Neural Tangent Kernels (NTKs) reveals that those long-tail data are commonly overlooked in the model's gradient updates and, consequently, are not effectively memorized, leading to poor domain-specific downstream performance. Based on the intuition that data with similar semantic meaning are closer in the embedding space, we devise a Cluster-guided Sparse Expert (CSE) layer to actively learn long-tail domain knowledge typically neglected in previous pretrained LMs. During pretraining, a CSE layer efficiently clusters domain knowledge together and assigns long-tail knowledge to designate extra experts. CSE is also a lightweight structure that only needs to be incorporated in several deep layers. With our training strategy, we found that during pretraining, data of long-tail knowledge gradually formulate isolated, "outlier" clusters in an LM's representation spaces, especially in deeper layers. Our experimental results show that only pretraining CSE-based LMs is enough to achieve superior performance than regularly pretrained-finetuned LMs on various downstream tasks, implying the prospects of domain-specific-pretraining-free language models. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Interpretable Drug Response Prediction through Molecule Structure-aware and Knowledge-Guided Visible Neural Network

Interpretable Drug Response Prediction through Molecule Stru...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Xie, Jiancong Zhang, Zhe Li, Youyou Rao, Jiahua Yang, Yuedong Sun Yat-sen University School of Computer Science and Engineering China University of Chicago Department of Chemistry IL United States Sun Yat-sen University Key Laboratory of Machine Intelligence and Advanced Computing of Moe China

ISBN: (纸本)9798350386226

Precise prediction of anti-cancer drug responses has become a crucial obstruction in anti-cancer drug design and clinical applications. In recent years, various deep learning methods have been applied to drug response prediction and become more accurate. However, they are still criticized as being non-transparent. To offer reliable drug response prediction in real-world applications, there is still a pressing demand to develop a model with high predictive performance as well as interpretability. In this study, we propose DrugVNN, an end-to-end interpretable drug response prediction framework, which extracts gene features of cell lines through a knowledge-guided visible neural network (VNN) and learns drug representation through a node-edge communicative message passing network (CMPNN). Additionally, between these two networks, a novel drug-aware gene attention gate is designed to direct the drug representation to VNN to simulate the effects of drugs. By evaluating on the GDSC dataset, DrugVNN achieved state-of-the-art performance. Moreover, DrugVNN can identify active genes and relevant signaling pathways for specific drug-cell line pairs with supporting evidence in the literature, implying the interpretability of our model. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Cloud Energy Storage Management Including Smart Home Physical Parameters

Cloud Energy Storage Management Including Smart Home Physica...

引用

2023 IEEE IAS Global Conference on Emerging Technologies, GlobConET 2023

作者： Saini, Vikash Kumar Yelisetti, Srinivas Kumar, Rajesh Al-Sumaiti, Ameena S. Malaviya National Institute of Technology Center for Energy and Environment Jaipur India Khalifa University Advanced Power and Energy Center Electrical Engineering and Computer Science United Arab Emirates Malaviya National Institute of Technology Department of Electrical Engineering Jaipur India

ISBN: (纸本)9798350331790

Consumption of green energy in residential communities is increasing compared to conventional supply. However, the variability in generation due to different weather parameters is a significant challenge to their growth rate. Energy storage has the potential to address this issue, and sharing economy-based cloud energy storage (CES) has gained popularity as a way to reduce energy consumption costs and increase revenue. This study analyzes the deployment of CES infrastructure operation in seven residential smart homes community. All seven homes utilized CES services based on their daily net load demand. In the first phase, the PSO algorithm optimized the load trajectory, taking into account all smart home parameters and physical components. Subsequently, the operational cost of CES in the community is evaluated in scenarios with and without home energy management systems. Simulation results indicated that the energy consumption cost of the residential community from the grid is zero, and revenue is 43.59% higher when home energy management systems are employed. The modeled home parameters significantly affected total community energy costs. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Enhancing Early Diagnosis of Autism Spectrum Disorder Using a Bayesian-Optimized Random Forest Algorithm

引用

Procedia computer science 2025年 258卷 1576-1585页

作者： Tina Babu Rekha R Nair Kishore S Sheila Mahapatra Department of Computer Science and Engineering School of Advanced Computing Alliance University Bengaluru India Department of Computer Applications Nitte University (Deemed to be University) Nitte Institute of Professional Education Mangalore India Department of Electrical and Electronics Engineering Alliance University Bangalore India

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition marked by social communication difficulties, re- stricted interests, and repetitive behaviors. Early and accurate diagnosis is critical for timely intervention and improved outcomes. Traditional diagnostic methods, which rely on clinical observations and assessments, can be subjective and time-consuming. This study introduces a novel approach to predicting ASD using a machine learning technique, specifically a Bayesian-optimized Ran- dom Forest classification algorithm. The proposed method uses a comprehensive dataset containing various behavioral, cognitive, and demographic features associated with ASD. The Optimized Random Forest with Bayesian approach is used to categorize indi- viduals into two groups: those who have ASD and those who do not. To develop this ensemble learning technique, multiple decision trees are used, and to achieve higher performance as well as better capability to generalize, the hyperparameters are optimized with Bayesian optimization. Preliminary analysis and comparison with existing databases substantiated the findings and revealed the effectiveness of the proposed approach. The Bayesian-optimized Random Forest classifier achieved an accuracy value of 0.9890. Such performance is highly promising and could enable increased use of machine learning methods in timely and accurate diagno- sis of ASD. The proposed approach provides recommendations for a verified diagnostic technique when implemented and creates a strong foundation for future investigations into the use of machine learning for identifying and supporting children with ASD.

关键词： Autism Spectrum Disorder (ASD) Machine Learning Bayesian Optimization Random Forest Early Diagnosis Predictive Modeling

来源：评论

学校读者我要写书评

暂无评论

Super-Resolution of Diffusion-Weighted Images via TDI-Conditioned Diffusion Model 15th

Super-Resolution of Diffusion-Weighted Images via TDI-Cond...

引用

15th International Workshop on Computational Diffusion MRI, CDMRI 2024, held in conjunction with 27th International Conference on Medical Image Computing and computer-Assisted Intervention, MICCAI 2024

作者： Ma, Jiquan Teng, Yujun Chen, Geng Jiang, Haotian Zhang, Kai Liu, Feihong Rekik, Islem Shen, Dinggang School of Computer Science and Technology Heilongjiang University Harbin China National Engineering Laboratory for Integrated Aero-Space-Ground-Ocean Big Data Application Technology School of Computer Science and Engineering Northwestern Polytechnical University Xi’an China School of Biomedical Engineering and State Key Laboratory of Advanced Medical Materials and Devices ShanghaiTech University Shanghai China School of Information Science and Technology Northwest University Xi’an China BASIRA Lab Imperial-X and Department of Computing Imperial College London London United Kingdom

ISBN: (纸本)9783031869198

Diffusion-Weighted Imaging (DWI) is a significant technique for studying white matter. However, it suffers from low-resolution obstacles in clinical settings. Post-acquisition Super-Resolution (SR) can enhance the resolution of DWIs and has gained increasing research interest in recent years. An advanced generative model, the Diffusion Model (DM), exhibits particularly promising performance in image SR. However, effective conditions are required to bootstrap the DM for DWI SR. To this end, we proposed the first DM-based DWI SR model with two effective conditions based on low-solution DWIs and Track Density Imaging (TDI) maps, which possess rich high-resolution prior knowledge Additionally, we consider another condition based on features from low-resolution DWIs. These two conditions are integrated into our model, which comprises three components: DWI Resolution Enhancer (DRE), DWI Feature Extractor (DFE), and TDI Feature Extractor (TFE). DRE combines low-resolution DWI features from DFE with TDI features from TFE to progressively generate high-resolution DWIs. We performed extensive experiments on DWIs of normal subjects from human connectome projects and patients with Parkinson’s disease. The results demonstrate that our model outperforms existing DWI SR models, both qualitatively and quantitatively. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Neurodegenerative diseases

来源：评论

学校读者我要写书评

暂无评论

Ultrahigh-coherence microcomb generation and low-noise amplification using integrated photonics

Ultrahigh-coherence microcomb generation and low-noise ampli...

引用

2024 IEEE Photonics Conference, IPC 2024

作者： Zhang, Xuguang Zhou, Zixuan Guo, Yijun Zhuang, Minxue Jin, Warren Shen, Bitao Chen, Yujun Huang, Jiahui Hu, Weiwei Shu, Haowen Bowers, John E. Wang, Xingjun Chang, Lin University of California Santa Barbara Department of Electrical and Computer Engineering Santa BarbaraCA United States Peking University State Key Laboratory of Advanced Optical Communications System and Networks School of Electronics Beijing China Peking University Frontiers Science Center for Nano-Optoelectronics Beijing China

ISBN: (纸本)9798350361957

We demonstrate a high-coherence parallel integrated light source utilizing a self-injection locked microcomb and injection locking amplification. Our strategy exhibits a record high on-chip gain of 60 dB, nearly 0 dB noise factor, which enable more than 30 highly coherent channels in integrated photonics whose linewidths down to 10 Hz, power exceeding 20 dBm per line and a wall plug efficiency of 19%. © 2024 IEEE.

关键词： Photonic integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Path-guided particle-based sampling 24

Path-guided particle-based sampling

引用

Proceedings of the 41st International Conference on Machine Learning

作者： Mingzhou Fan Ruida Zhou Chao Tian Xiaoning Qian Department of Electrical & Computer Engineering Texas A&M University College Station Texas Department of Electrical and Computer Engineering University of California Los Angeles CA Department of Electrical & Computer Engineering and Department of Computer Science and Engineering Texas A&M University College Station TX and Computational Science Initiative Brookhaven National Laboratory Upton NY

Particle-based Bayesian inference methods by sampling from a partition-free target (posterior) distribution, e.g., Stein variational gradient descent (SVGD), have attracted significant attention. We propose a path-guided particle-based sampling (PGPS) method based on a novel Log-weighted Shrinkage (LwS) density path linking an initial distribution to the target distribution. We propose to utilize a Neural network to learn a vector field motivated by the Fokker-Planck equation of the designed density path. Particles, initiated from the initial distribution, evolve according to the ordinary differential equation defined by the vector field. The distribution of these particles is guided along a density path from the initial distribution to the target distribution. The proposed LwS density path allows for an efficient search of modes of the target distribution while canonical methods fail. We theoretically analyze the Wasserstein distance of the distribution of the PGPS-generated samples and the target distribution due to approximation and discretization errors. Practically, the proposed PGPS-LwS method demonstrates higher Bayesian inference accuracy and better calibration ability in experiments conducted on both synthetic and real-world Bayesian learning tasks, compared to baselines, such as SVGD and Langevin dynamics, etc.

关键词：

来源：评论

学校读者我要写书评

暂无评论

LIAS: A Lightweight Incentive Authentication Scheme for Forensic Services in IoV

LIAS: A Lightweight Incentive Authentication Scheme for Fore...

引用

作者： Zhang, Mingyue Zhou, Junlong Cong, Peijin Zhang, Gongxuan Zhuo, Cheng Hu, Shiyan Nanjing University of Science and Technology School of Computer Science and Engineering Nanjing210094 China Institute of Computing Technology Chinese Academy of Sciences State Key Laboratory of Computer Architecture Beijing100190 China Zhejiang University Department of Information Science and Electronic Engineering Hangzhou310027 China University of Southampton School of Electronics and Computer Science SouthamptonSO17 1BJ United Kingdom

Internet of Vehicles (IoV) has become an indispensable data sensing and processing platform in Internet of Things (IoT) for intelligent transportation. The mounted cameras on the vehicles along with the fixed roadside cameras are utilized to provide pictorial services for IoV users and law enforcement agencies. For such forensic services, ensuring the security and privacy of vehicles while guaranteeing the efficiency of data transmission among vehicles is important. In this paper, we propose a lightweight incentive authentication scheme (LIAS) for forensic services in IoV. LIAS is developed on a three-tier architecture containing cloud layer, fog layer, and user layer. LIAS uses pairing-free certificateless signcryption, pseudonym update mechanism, and incentive mechanism to realize a secure anonymous authentication efficiently. We conduct correctness and security analysis, as well as performance analysis and evaluation to validate the high security and efficiency of LIAS. Experimental results reveal that, the communication and computation overheads as well as the message delay and packet loss of LIAS are much lower than those of state-of-the-art techniques. Note to Practitioners - This paper is motivated by the security and privacy issues of forensic services in IoV for intelligent transportation. Our goal is to improve the security and privacy of vehicles while guaranteeing the lightweight and incentive of data transmission among the vehicles. Fog-assisted IoV is introduced to fully utilize the capacities of near-user edge devices as well as the connections between fog nodes and devices. However, it still faces the difficulties in ensuring vehicles' security and privacy. Moreover, vehicles' information dissemination could be easily monitored because of the unavoidable defect of wireless communication. Thereby, it is essential to guarantee the security and privacy of vehicles while enhancing the efficiency of vehicles' data transmission during the forensic service.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Visual SLAM with a Multi-Modal Semantic Framework for the Visually Impaired Navigation-Aided Device 8

Visual SLAM with a Multi-Modal Semantic Framework for the Vi...

引用

8th IEEE International Conference on advanced Robotics and Mechatronics, ICARM 2023

作者： Li, Shaohu Gu, Jason Feng, Ying Institute of Advanced Technology University of Science and Technology of China Hefei230031 China Dalhousie University Department of Electrical and Computer Engineering HalifaxNS Canada College of Automation Science and Engineering South China University of Technology Guangzhou510641 China

ISBN: (纸本)9798350300178

Visual Simultaneous Localization and Mapping (VSLAM) based on image feature points often contains feature points on non-stationary objects in dynamic scenes, resulting in poor localization performance. Recent research has improved VSLAM performance by removing dynamic feature points utilizing optical flow, deep learning, or multi-view geometry, however, the volume of data makes real-time operation challenging. In this study, we propose a multimodal framework for semantic VSLAM. To eliminate dynamic feature points, the system effectively blends object detection, instance segmentation, and geometry modules. To ensure high-precision pose estimation and quick back-end optimization, we exclusively use object detection in the front-end of VSLAM to mask dynamic feature points. In the keyframes of the local map, semantic segmentation is utilized to remove dynamic feature points, and an effective geometric module is employed to assess the dynamic consistency of objects that are challenging to categorize. Therefore, more feature points are retained, and the running speed is also ensured. Evaluations of the system on public datasets and in the real world show that it improves the accuracy and stability of VSLAM in dynamic environments. © 2023 IEEE.

关键词： Object detection

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：