检索结果-内蒙古大学图书馆

SSRN 2023年

作者： Agarwal, Anish Dahleh, Munther Horel, Thibaut Rui, Maryann Institute for Data Systems and Society Massachusetts Institute of Technology CambridgeMA02139 United States Department of Electrical Engineering and Computer Science

The design of data markets has gained importance as firms increasingly use machine learning models fueled by externally acquired training data. A key consideration is the externalities firms face when data, though inherently freely replicable, is allocated to competing firms. In this setting, we demonstrate that a data seller’s optimal revenue increases as firms can pay to prevent allocations to others. To do so, we first reduce the combinatorial problem of allocating and pricing multiple datasets to the auction of a single digital good by modeling utility for data through the increase in prediction accuracy it provides. We then derive welfare and revenue maximizing mechanisms, highlighting how the form of firms’ private information– whether the externalities one exerts on others is known, or vice-versa– affects the resulting structures. In all cases, the optimal allocation rule is a single threshold per firm, where either all data is allocated or none is. © 2023, The Authors. All rights reserved.

关键词： Machine design

来源：评论

学校读者我要写书评

暂无评论

Improving routing efficiency in vehicular AD Hoc Networks

Improving routing efficiency in vehicular AD Hoc Networks

引用

International Conference on Control, Decision and Information Technologies (CoDIT)

作者： W. Akermi N. Alyaoui A. B. Guiloufi K. Chabbir N. Nasri Automatic Industrial Systems Department Applied Science and Technology Gafsa University Gafsa Tunisia Laboratory of Modeling Analysis and Control of Systems (MACS) National Engineering School of Gabes (ENIG) Gabes University Gabes Tunisia Computer Science and Telecommunications Department Higher Institute of Applied Science and Technology Gafsa University Gafsa Tunisia Preparatory Institute for Engineering Studies of Gabes Laboratory of Modeling Analysis and Control of Systems (MACS) National Engineering School of Gabes (ENIG) Gabes University Gabes Tunisia Computer Science Department-Gafsa Gafsa University Gafsa Tunisia Laboratory of Smart Systems for Engineering and E-Health Based on Technologies of Image and Telecommunications (SETIT) Sfax University Sfax Tunisia

ISBN: (数字)9798350373974

ISBN: (纸本)9798350373981

Vehicular networks, also known as Vehicular Ad Hoc Networks (VANETs), have become a crucial technology for vehicle communication, providing wireless connectivity among vehicles and between vehicles and infrastructure. However, the dynamic and complex nature of these networks presents unique challenges in routing and resource management. In this study, we explore the use of artificial intelligence (AI) to improve the performance of vehicular networks, focusing on two widely used routing protocols: Optimized Link State Routing (OLSR) and Destination-Sequenced Distance Vector (DSDV).We aim to evaluate the performance of these traditional routing protocols within the context of vehicular networks by integrating machine learning techniques such as neural network regression (MLRegression). Our goal is to understand how the integration of artificial intelligence can enhance the prediction and optimization of vehicular network performance. Using MLRegression, we aim to develop accurate prediction models for performance metrics such as latency, throughput, packet delivery rate, and more, based on network parameters and environmental conditions. This comparative evaluation, which involves incorporating an AI-based approach, will allow us to better understand the strengths and weaknesses of each protocol and provide recommendations for improving vehicular networks.

关键词： Measurement Wireless communication Adaptation models Adaptive systems Vehicular ad hoc networks Routing Throughput Routing protocols Vectors Vehicle dynamics

来源：评论

学校读者我要写书评

暂无评论

Autonomous Search and Rescue Agent Using Deep Reinforcement Learning

Autonomous Search and Rescue Agent Using Deep Reinforcement ...

引用

Recent Advances in science and engineering Technology (ICRASET), International Conference on

作者： R. S. Rampriya Mohammed Junaid Alam Sri Jayan School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore Tamilnadu India Department of Computer Technology Madras Institute of Technology Chennai Tamilnadu India

Disaster management is one of the areas of paramount importance for any government or organization. This process gets very tedious and tricky for some specific type of disasters. Deploying manpower for mitigating the disaster and saving the victims is a risky decision because the life of the people in the disaster management team is put to risk, which is a very unfortunate thing. So instead of using manpower, we can deploy intelligent machine agents, which are equivalent or sometimes more efficient than human beings for such highly risky tasks and avert the loss of lives. In our case, we have specifically considered the disaster wherein fire breaks out in a building and people get trapped inside due to the dangerous situation. In this case, where deploying manpower to rescue them requires a high quality protective gear and can even put their lives on risk, we have proposed to deploy intelligent machine agents that are trained to figure out the path on their own and find their way to the victims after which a rescue operation can be performed by safely escorting the victims to a nearby exit in the building. The agent in the proposed application has been evaluated by the metrics such as the time taken to reach the victims and come back to the exit, the total rewards gathered by the agent during its exploration in the environment and the number of correct decisions it made during its journey. The agent takes an average time of 145.6 seconds to complete its operation in the real-time environment. It gathered an average peak cumulative reward value of -0.2 in the real time environment. The agent on an average, records 67.3% chance of making a correct decision (that gets a positive reward) given a state vector.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Career Track Based Capstone Course Design for Information systems Curriculum

Career Track Based Capstone Course Design for Information Sy...

引用

IEEE Education engineering (EDUCON)

作者： Swapna Gottipati Kyong Jin Shim Subhajit Datta School of Computing and Information Systems Singapore Management University Singapore Department of Computer Science and Engineering Heritage Institute of Technology Kolkata India

ISBN: (数字)9798350399431

ISBN: (纸本)9798350399448

The aim of technical education is to design a career-based curriculum that imparts a set of skills to students and enable them to contribute to a nation's economic progress. With their multifarious uses, information systems (IS) lie at the heart of modern economies. IS curricula offer several tracks for students to pursue career paths based on their interests and opportunities. To offer students an opportunity to apply the knowledge they glean from classroom instructions on real-world problems, capstone projects are essential elements in most, if not all, IS programs. As IS programs increasingly offer a wider range of career paths for students, the design and delivery of such capstone projects present a set of unique challenges. In this paper, we examine the set of challenges and share our experience from designing and delivering a career track-based capstone course for the first time at a large university. Feedback from students and faculty supervisors indicates that the course was able to meet its learning objectives to a notable extent and offered important pointers to future improvements in the course design.

关键词： Heart Economics engineering profession Stakeholders engineering education Information systems

来源：评论

学校读者我要写书评

暂无评论

Generative De-Quantization for Neural Speech Codec Via Latent Diffusion

Generative De-Quantization for Neural Speech Codec Via Laten...

引用

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

作者： Haici Yang Inseon Jang Minje Kim Department of Intelligent Systems Engineering Indiana University Bloomington IN USA Electronics and Telecommunications Research Institute Daejeon South Korea Department of Computer Science University of Illinois at Urbana-Champaign IL USA

End-to-end speech coding models achieve high coding gains by learning compact yet expressive features and a powerful decoder in a single network. A challenging problem as such results in unwelcome complexity increase and inferior speech quality. In this paper, we propose to separate the representation learning and information reconstruction tasks. We leverage an end-to-end codec for learning low-dimensional discrete tokens. Instead of using its decoder, we employ a latent diffusion model to de-quantize coded features into a high-dimensional continuous space, relieving the decoder’s burden of de-quantizing and upsampling. To mitigate the issue of over-smooth generation, we introduce midway-infilling with less noise reduction and stronger conditioning. We investigate the hyperparameters for midway-infilling and latent diffusion space with different dimensions in ablation studies. Subjective listening tests show that our model outperforms the state-of-the-art at two low bitrates, 1.5 and 3 kbps. We open-source the project for reproducibility 1 .

关键词：

来源：评论

学校读者我要写书评

暂无评论

Active Large Language Model-based Knowledge Distillation for Session-based Recommendation

arXiv

引用

arXiv 2024年

作者： Du, Yingpeng Sun, Zhu Wang, Ziyan Chua, Haoyan Zhang, Jie Ong, Yew-Soon College of Computing and Data Science Nanyang Technological University Singapore Information Systems Technology and Design Singapore University of Technology and Design Singapore Department of Computer Science and Engineering Yonsei University Korea Republic of

Large language models (LLMs) provide a promising way for accurate session-based recommendation (SBR), but they demand substantial computational time and memory. Knowledge distillation (KD)-based methods can alleviate these issues by transferring the knowledge to a small student, which trains a student based on the predictions of a cumbersome teacher. However, these methods encounter difficulties for LLM-based KD in SBR. 1) It is expensive to make LLMs predict for all instances in KD. 2) LLMs may make ineffective predictions for some instances in KD, e.g., incorrect predictions for hard instances or similar predictions as existing recommenders for easy instances. In this paper, we propose an active LLM-based KD method in SBR, contributing to sustainable AI. To efficiently distill knowledge from LLMs with limited cost, we propose to extract a small proportion of instances predicted by LLMs. Meanwhile, for a more effective distillation, we propose an active learning strategy to extract instances that are as effective as possible for KD from a theoretical view. Specifically, we first formulate gains based on potential effects (e.g., effective, similar, and incorrect predictions by LLMs) and difficulties (e.g., easy or hard to fit) of instances for KD. Then, we propose to maximize the minimal gains of distillation to find the optimal selection policy for active learning, which can largely avoid extracting ineffective instances in KD. Experiments on real-world datasets show that our method significantly outperforms state-of-the-art methods for SBR. Copyright © 2024, The Authors. All rights reserved.

关键词： Active learning

来源：评论

学校读者我要写书评

暂无评论

Graph Kolmogorov-Arnold Networks for Multi-Cancer Classification and Biomarker Identification, An Interpretable Multi-Omics Approach

arXiv

引用

arXiv 2025年

作者： Alharbi, Fadi Budhiraja, Nishant Vakanski, Aleksandar Zhang, Boyu Elbashir, Murtada K. Mohammed, Mohanad College of Engineering Department of Computer Science University of Idaho MoscowID83844 United States College of Computer and Information Sciences Department of Computer Science Jouf University Aljouf Sakaka72441 Saudi Arabia College of Computer and Information Sciences Department of Information Systems Jouf University Aljouf Sakaka72441 Saudi Arabia Faculty of Mathematical and Computer Sciences University of Gezira Wad Madani Sudan School of Mathematics Statistics and Computer Science University of KwaZulu-Natal Scottsville Pietermaritzburg3209 South Africa

The integration of multi-omics data presents a major challenge in precision medicine, requiring advanced computational methods for accurate disease classification and biological interpretation. This study introduces the Multi-Omics Graph Kolmogorov–Arnold Network (MOGKAN), a deep learning model that integrates messenger-RNA, micro-RNA sequences, and DNA methylation data with Protein-Protein Interaction (PPI) networks for accurate and interpretable cancer classification across 31 cancer types. MOGKAN employs a hybrid approach combining differential expression with DESeq2, Linear Models for Microarray (LIMMA), and Least Absolute Shrinkage and Selection Operator (LASSO) regression to reduce multi-omics data dimensionality while preserving relevant biological features. The model architecture is based on the Kolmogorov–Arnold theorem principle, using trainable univariate functions to enhance interpretability and feature analysis. MOGKAN achieves classification accuracy of 96.28% and demonstrates low experimental variability with a standard deviation that is reduced by 1.58–7.30% points compared to Convolutional Neural Networks (CNNs) and Graph Neural Networks (GNNs). The biomarkers identified by MOGKAN have been validated as cancer-related markers through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis. The proposed model presents an ability to uncover molecular oncogenesis mechanisms by detecting phosphoinositide-binding substances and regulating sphingolipid cellular processes. By integrating multi-omics data with graph-based deep learning, our proposed approach demonstrates superior predictive performance and interpretability that has the potential to enhance the translation of complex multi-omics data into clinically actionable cancer *** Codes 68 Copyright © 2025, The Authors. All rights reserved.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Invertibility of Discrete-Time Linear systems with Sparse Inputs

arXiv

引用

arXiv 2024年

作者： Poe, Kyle Mallada, Enrique Vidal, René Applied Mathematics and Computational Science Group University of Pennsylvania PA19104 United States Department of Electrical and Computer Engineering Johns Hopkins University MD21218 United States Departments of Electrical and Systems Engineering Radiology Computer and Information Science Statistics and Data Science University of Pennsylvania PA19104 United States

One of the fundamental problems of interest for discrete-time linear systems is whether its input sequence may be recovered given its output sequence, a.k.a. the left inversion problem. Many conditions on the state space geometry, dynamics, and spectral structure of a system have been used to characterize the well-posedness of this problem, without assumptions on the inputs. However, certain structural assumptions, such as input sparsity, have been shown to translate to practical gains in the performance of inversion algorithms, surpassing classical guarantees. Establishing necessary and sufficient conditions for left invertibility of systems with sparse inputs is therefore a crucial step toward understanding the performance limits of system inversion under structured input assumptions. In this work, we provide the first necessary and sufficient characterizations of left invertibility for linear systems with sparse inputs, echoing classic characterizations for standard linear systems. The key insight in deriving these results is in establishing the existence of two novel geometric invariants unique to the sparse-input setting, the weakly unobservable and strongly reachable subspace arrangements. By means of a concrete example, we demonstrate the utility of these characterizations. We conclude by discussing extensions and applications of this framework to several related problems in sparse control. © 2024, CC BY.

关键词： Linear systems

来源：评论

学校读者我要写书评

暂无评论

Ether Vote: Revolutionizing Elections with Blockchain-Powered Electronic Voting System

Ether Vote: Revolutionizing Elections with Blockchain-Powere...

引用

Smart Generation Computing, Communication and Networking (SMART GENCON), International Conference on

作者： Sijo Joseph Priyank Pandey Manju Khari Kapil Kumar Piyush Pratap Singh School of Computer & Systems Sciences Jawaharlal Nehru University New Delhi India Department of Computer Science & Engineering Graphic era deemed to be university Dehradun Uttarakhand India

Voting is a crucial part of democracy because it gives people the chance to voice their opinions, hold elected officials responsible, encourage diversity in the government, cultivate civic participation, and defend themselves from tyranny. Existing electronic voting (e-voting) systems do in fact suffer a number of important obstacles, with security difficulties and a lack of transparency ranking as two of the most important problems. Given the significance of elections in democracies and the likelihood of fraud or other forms of manipulation in electronic voting procedures, these issues are extremely pressing. systems used for electronic voting heavily rely on software, which might occasionally have vulnerabilities that hackers can exploit. Any weak point in the system can be used to sway elections or jeopardize its security. To address these issues, a strong, secure electronic voting system (EVS), an open system design, impartial audits, and a dedication to inclusion and accessibility are required. For democratic processes to continue to be trusted and the right to vote to be protected, e-voting system integrity must be ensured. It is necessary to create a new Electronic Voting System (EVS) that can offer greater security, speed, and accuracy than the EVS used in the past. The authors of this research suggested a blockchain-based secure EVS. Immutable, transparent, and secure distributed ledger technology named as blockchain. Blockchain used to implement an E-Voting system that is transparent, tamper-proof, and can guarantee the correctness and integrity of the voting process.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Interference reduction in directional wireless networks 17th

Interference reduction in directional wireless networks

引用

17th International Conference on Distributed Computing and Internet Technology, ICDCIT 2021

作者： Basappa, Manjanna Mishra, Sudeepta Department of Computer Science and Information Systems Birla Institute of Technology and Science Pilani Hyderabad Campus Shameerpet500078 India Department of Computer Science and Engineering Indian Institute of Technology Ropar Nangal Road Rupnagar140001 India

ISBN: (纸本)9783030656201

In a wireless network using directional transmitters, a typical problem is to schedule a set of directional links to cover all the receivers in a region, such that an adequate data rate and coverage are maintained while minimizing interference. We can model the coverage area of a directional transmitter as an unit triangle and the receiver as a point in the plane. Motivated by this, we first consider a minimum ply covering (MPC) problem. We propose a 2-approximation algorithm for the MPC problem in O((opt+ n) m14opt+1(log opt)) time, where m is the number of transmitters and n is the number of receivers given in the plane, and opt is the maximum number of triangles, in the optimal solution, covering a point in the plane. We also show that the MPC problem is NP-hard, and is not (2 - ϵ) -approximable for any ϵ> 0 unless P = NP. We also study channel allocation in directional wireless networks by posing it as a colorable covering problem, namely, 3-colorable unit triangle cover (3CUTC). We propose a simple 4-approximation algorithm in O(m30n2) time, for this problem. © Springer Nature Switzerland AG 2021.

关键词： Approximation algorithms

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：