检索结果-内蒙古大学图书馆

A Fractional Sir Model for the Simulation of the Spread of Influenza a During the Post Covid-19 Period

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Huang, Jian Cen, Zhongdi Xu, Aimin He, Tianfeng Liu, Qingbing School of Big Data and Software Engineering Zhejiang Wanli University Ningbo315100 China Ningbo Municipal Center for Disease Control and Prevention Ningbo315010 China

In this article the fractional SIR model is used to simulate the spread of influenza A in Ningbo from January 1, 2023 to June 30, 2023. The $L1$ method and nonstandard finite difference method are used to discretize the nonlinear fractional SIR model, which leads to the derivation of an explicit formulation. It is proved that the explicit scheme preserve the positivity and asymptotical stability. Under limited data conditions the model parameter fitting problem is transformed into an optimization problem. The optimization problem transformed from the fractional SIR model is solved by the improved QPSO algorithm. Numerical results show that the goodness of fit for the fractional SIR model is 0.941 and the fractional SIR model can well reflect the actual spread characteristics of influenza A during the post COVID-19 period. © 2024, The Authors. All rights reserved.

关键词： COVID-19

Mellivora Capensis: A Backdoor-Free Training Framework on the Poisoned dataset without Auxiliary data

学校读者我要写书评

暂无评论

arXiv

arXiv 2024年

作者： Pu, Yuwen Chen, Jiahao Zhou, Chunyi Feng, Zhou Li, Qingming Hu, Chunqiang Ji, Shouling The College of Computer Science and Technology Zhejiang University Zhejiang Hangzhou310027 China The School of Big Data & Software Engineering Chongqing University Chongqing400030 China

The efficacy of deep learning models is profoundly influenced by the quality of their training data. Given the considerations of data diversity, data scale, and annotation expenses, model trainers frequently resort to sourcing and acquiring datasets from online repositories. Although economically pragmatic, this strategy exposes the models to substantial security vulnerabilities. Untrusted entities can clandestinely embed triggers within the dataset, facilitating the hijacking of the trained model on the poisoned dataset through backdoor attacks, which constitutes a grave security concern. Despite the proliferation of countermeasure research, their inherent limitations constrain their effectiveness in practical applications. These include the requirement for substantial quantities of clean samples, inconsistent defense performance across varying attack scenarios, and inadequate resilience against adaptive attacks, among others. Therefore, in this paper, we endeavor to address the challenges of backdoor attack countermeasures in real-world scenarios, thereby fortifying the security of training paradigm under the data-collection manner. Concretely, we first explore the inherent relationship between the potential perturbations and the backdoor trigger, and demonstrate the key observation that the poisoned samples perform more robustness to perturbation than the clean ones through the theoretical analysis and experiments. Then, based on our key explorations, we propose a robust and clean-data-free backdoor defense framework, namely Mellivora Capensis (MeCa), which enables the model trainer to train a clean model on the poisoned dataset. MeCa can effectively identify poisoned samples and train a clean model on the poisoned dataset and eliminate the requirement of clean samples or knowledge about the poisoned dataset (e.g., poisoning ratio). We conduct extensive experiments in defending against 8 SOTA attacks (including 3 adaptive attacks) on 4 datasets (Imagnette, Tiny I

关键词： Adversarial machine learning

Tkstagnet: A Top-K Spatio-Temporal Attention Gating Network for Air Pollution Prediction

学校读者我要写书评

暂无评论

SSRN 2024年

作者： Zhou, Hongliang Mao, Yingchi Li, Xiang Rong, Yi Chen, Ling Yin, Changkui College of Computer Science and Software Engineering Hohai University Nanjing China Key Laboratory of Water Big Data Technology of Ministry of Water Resources Nanjing China

Accurate air pollution prediction is of great significance for human health, travel and environmental management. As an emerging artificial intelligence technology, Transformer has already shown tremendous potential in air pollution prediction. Unfortunately, existing Transformer-based methods are challenging to accurately predict air pollution due to two limitations: 1) they neglect irrelevant spatial and temporal neighbors (e.g., neighboring sensors and timesteps) when modeling spatio-temporal correlations, and 2) they lack an adaptive mechanism to mine different effects caused by various external factors (e.g., meteorological conditions). Jointly taking these limitations into account, we propose a novel Transformer-based architecture for air pollution prediction, named Top-K Spatio-Temporal Attention Gating Network (TKSTAGNet). In this architecture, to effectively extract spatio-temporal correlations, we first design a TopKpooling Spatial Attention (TopK-SA) and a TopKpooling Temporal Attention (TopK-TA). In TopK-SA, the TopKpooling mechanism is introduced into spatial attention to filter out irrelevant spatial neighbors, accurately capturing spatial dependencies. In TopK-TA, the TopKpooling mechanism is integrated with temporal attention to identify and exclude irrelevant temporal neighbors, precisely extracting temporal dependencies. To adaptively fuse various external factors, a gating fusion module is then proposed. We conduct experiments on two real-world air pollution datasets from urban regions in Beijing and Shanghai, respectively. Experimental results demonstrate that TKSTAGNet outperforms state-of-the-art baselines in prediction precision and fitting the variation trends of air pollution. © 2024, The Authors. All rights reserved.

关键词： Air pollution

Requirements engineering for Machine Learning: A Review and Reflection

学校读者我要写书评

暂无评论

arXiv

arXiv 2022年

作者： Pei, Zhongyi Liu, Lin Wang, Chen Wang, Jianmin National Engineering Research Center for Big Data Software School of Software Tsinghua University Beijing China

Today, many industrial processes are undergoing digital transformation, which often requires the integration of well-understood domain models and state-of-the-art machine learning technology in business processes. However, requirements elicitation and design decision making about when, where and how to embed various domain models and end-to-end machine learning techniques properly into a given business workflow requires further exploration. This paper aims to provide an overview of the requirements engineering process for machine learning applications in terms of cross domain collaborations. We first review the literature on requirements engineering for machine learning, and then go through the collaborative requirements analysis process step-by-step. An example case of industrial data-driven intelligence applications is also discussed in relation to the aforementioned steps. © 2022, CC BY.

关键词： Requirements engineering

Dullahan: Stealthy Backdoor Attack against Without-Label-Sharing Split Learning

学校读者我要写书评

暂无评论

arXiv

arXiv 2024年

作者： Pu, Yuwen Ding, Zhuoyuan Chen, Jiahao Zhou, Chunyi Li, Qingming Hu, Chunqiang Ji, Shouling The College of Computer Science and Technology Zhejiang University Zhejiang Hangzhou310027 China The School of Big Data & Software Engineering Chongqing University Chongqing400030 China

As a novel privacy-preserving paradigm aimed at reducing client computational costs and achieving data utility, split learning has garnered extensive attention and proliferated widespread applications across various fields, including smart health and smart transportation, among others. While recent studies have primarily concentrated on addressing privacy leakage concerns in split learning, such as inference attacks and data reconstruction, the exploration of security issues (e.g., backdoor attacks) within the framework of split learning has been comparatively limited. Nonetheless, the security vulnerability within the context of split learning is highly posing a threat and can give rise to grave security implications, such as the illegal impersonation in the face recognition model. Therefore, in this paper, we propose a Stealthy Backdoor Attack Strategy (namely Dullahan) tailored to the without-label-sharing split learning architecture, which unveils the inherent security vulnerability of split learning. We posit the existence of a potential attacker on the server side aiming to introduce a backdoor into the training model, while exploring two scenarios: one with known client network architecture and the other with unknown architecture. Diverging from traditional backdoor attack methods that manipulate the training data and labels, we constructively conduct the backdoor attack by injecting the trigger embedding into the server network. Specifically, our Dullahan achieves a higher level of attack stealthiness by refraining from modifying any intermediate parameters (e.g., gradients) during training and instead executing all malicious operations post-training. Finally, we conducted extensive experiments on 3 different models, 3 datasets, and various splitting strategies, and the results demonstrate that our approach achieves a considerable attack success rate while causing minimal impact on the main task's performance. In this paper, we unveil the inherent security v

关键词： Adversarial machine learning

A Multiple Kernel-Based Kernel Density Estimator for Multimodal Probability Density Functions

学校读者我要写书评

暂无评论

SSRN 2023年

作者： He, Yu-Lin Chen, Jia-Qi Fournier-Viger, Philippe Huang, Joshua Zhexue Big Data Institute College of Computer Science & Software Engineering Shenzhen University Shenzhen518060 China Shenzhen518107 China

To approximate multimodal PDFs, which are frequently encountered in real-world applications, this paper proposes a novel multiple kernel-based kernel density estimator (MK-KDE). It constructs a flexible KDE using the weighted average of multiple kernels by considering their kernel efficiencies. To achieve this, a novel efficient objective function is designed to obtain the optimized kernel weights and kernel bandwidths by minimizing both the global estimation error of MK-KDE and the local estimation errors of SK-KDEs. Moreover, a sophisticated $k$-nearest neighbor strategy is devised as a heuristic method to determine the unknown PDF values of the given data points, thereby optimizing the aforementioned objective function. A series of persuasive experiments were conducted to validate the feasibility, rationality, and effectiveness of MK-KDE for 10 multimodal PDFs. The experimental results demonstrated that MK-KDE is a viable approach to estimate multimodal PDFs. © 2023, The Authors. All rights reserved.

关键词： k-nearest neighbors

A Novel Certificateless Authentication and Key Agreement Protocol for Smart Grid

学校读者我要写书评

暂无评论

A Novel Certificateless Authentication and Key Agreement Pro...

IEEE Conference on Global Communications (GLOBECOM)

作者： Zewei Liu Chunqiang Hu Conghao Ruan Ruinian Li School of Big Data and Software Engineering Chongqing University China Computer Science Department Bowling Green State University USA

With the rapid development of the smart grid (SG), the security and reliability of communication for SG has become an urgent and critical issue. The authentication and key agreement (AKA) protocol has emerged as important means to address the security challenges in smart grid. This paper proposes a novel certificateless authentication and key agreement protocol (NCL-AKA) for the smart grid, which overcomes the limitations of existing AKA protocol such as complex certificate management, untimely certificate revocation, and man-in-the-middle attacks. Firstly, the system architecture and threat model of the proposed protocol is presented. Then, the correctness and security of the proposed protocol are analyzed. Finally, comparative experiments and analysis reveal that the NCL-AKA has lower computational cost and communication overhead.

关键词：

DHFM-FLM: A Dynamic Hierarchical Federated Learning Mechanism for Financial Models under Client Resource Heterogeneity

学校读者我要写书评

暂无评论

DHFM-FLM: A Dynamic Hierarchical Federated Learning Mechanis...

2024 IEEE International Conference on big data, bigdata 2024

作者： Yin, Kangning Ding, Zhen Hou, Shaoqi Ji, Xinhui Yin, Guangqiang Wang, Zhiguo University of Electronic Science and Technology of China Kash Institute of Electronics and Information Industry Chengdu China School of Information Management Wuhan University Wuhan China Big Data Research Institute Wuhan University Wuhan China School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China Shanghai New Energy Vehicle Public Data Collection and Monitoring Research Center Shanghai China

ISBN: (纸本)9798350362480

Federated Learning (FL) is an emerging distributed machine learning technology. However, in practical applications, it frequently encounters the challenge of client resource heterogeneity. This can result in long wait times or even model training failures during the communication process of FL. To address this problem, we propose a dynamic hierarchical federated learning mechanism for financial models (DHFM-FLM). The local client adopts a dynamic model training design that leverages the property of resource heterogeneity to enhance the performance of the local model. To avoid prolonged wait times for failing clients, a dynamic communication detection design is proposed at three critical junctures. In each round, the model hierarchical reservation communication design is employed to collect models in segments, thus reducing communication congestion and preventing malicious attacks on the communication process. Experiments with heterogeneous computation and communication resources demonstrate that utilizing the DHFM-FLM boosts model performance by approximately 5-8% and reduces communication time by about 15%. Additionally, DHFM-FLM increases the success rate of the FL task by approximately 6%. © 2024 IEEE.

关键词： Federated learning

Higher Order Belief Divergence with its Application in Pattern Classification

学校读者我要写书评

暂无评论

SSRN 2022年

作者： Huang, Yingcheng Xiao, Fuyuan School of Big Data and Software Engineering Chongqing University Chongqing401331 China

Multi-source information fusion is a sophisticated estimation process that generates a unified profile to assess complex situations. Dempster–Shafer evidence theory (DSET) is a practical theory in handling uncertain information in multi-source information fusion. However, highly-conflicting evidence may cause combination rule of Dempster to provide counterintuitive results. Thus, how to effectively reconcile highly-conflicting evidence in DSET is still an open issue. To address this problem, in this paper, a novel belief divergence, higher order belief Jensen-Shannon divergence is proposed to measure the discrepancy between BPAs in DSET. The proposed higher order belief Jensen-Shannon divergence is the first method to dynamically measure the discrepancy between BPAs over the time evolution, i.e., to measure the discrepancy between BPAs with different time scale in the future. Besides, the proposed higher order belief Jensen-Shannon divergence has benefits from the perspective of measurement. It satisfies the properties of nonnegativeness and nondegeneracy, symmetry, and the triangle inequality of root form. Based on the proposed higher order belief Jensen-Shannon divergence, a novel multi-source information fusion algorithm is proposed. Eventually, the proposed algorithm is applied to a pattern classification experiment with real-world datasets. © 2022, The Authors. All rights reserved.

关键词： Classification (of information)