检索结果-内蒙古大学图书馆

3rd International Conference on Machine Learning and Data engineering, ICMLDE 2024

作者： Soundarya, B.C. Gururaj, H.L. Naveen Kumar, C.M. Department of Information Technology Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India Department of Computer Science and Business System Malnad College of Engineering Hassan India

With the advancement of Artificial Intelligence, facial recognition has become a crucial biometric feature. Deepfake technology leverages AI and can create hyper-realistic digitally manipulated videos of people appearing to say or do things that never occurred. The emergence of Generative Adversarial Networks (GANs) has further enabled the creation of fake visual content with astonishing realism. This technology has diverse applications, such as in the film industry, where it allows for video recreation without reshooting, creating awareness videos, restoring the voices of those who have lost them, and updating movie scenes at low cost. However, this rapid advancement also presents significant challenges. The proliferation of synthetic images raises severe concerns about their societal impact, particularly in terms of potential misuse for harassment and blackmail. Therefore, developing robust deepfake detection models is imperative. This study evaluates the performance of a proposed ResNet34 model in deepfake detection. We utilize the FaceForensics++ dataset to train and assess the model, incorporating images generated by four popular deepfake techniques. Our experimental results demonstrate that integrating linear ternary patterns (LTP) and edge detection-based features with the modified ResNet34 model achieves superior performance, attaining 97.5% accuracy and surpassing other approaches. © 2024 The Authors. Published by ELSEVIER B.V.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

MCTensor: A High-Precision Deep Learning Library with Multi-Component Floating-Point

arXiv

引用

arXiv 2022年

作者： Yu, Tao Guo, Wentao Li, Jianan Canal Yuan, Tiancheng De Sa, Christopher Department of Computer Science Biological Engineering System Engineering Cornell University United States

In this paper, we introduce MCTensor, a library based on PyTorch for providing general-purpose and high-precision arithmetic for DL training. MCTensor is used in the same way as PyTorch Tensor: we implement multiple basic, matrix-level computation operators and NN modules for MCTensor with identical PyTorch interface. Our algorithms achieve high precision computation and also benefits from heavily-optimized PyTorch floating-point arithmetic. We evaluate MCTensor arithmetic against PyTorch native arithmetic for a series of tasks, where models using MCTensor in float16 would match or outperform the PyTorch model with float32 or float64 precision. © 2022, CC BY.

关键词： Digital arithmetic

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence-based scheduling with Multi-Objective Power Optimization in Smart Homes Utilizing Particle Swarm Optimizer

Artificial Intelligence-based scheduling with Multi-Objectiv...

引用

2023 International Conference for Technological engineering and its Applications in Sustainable Development, ICTEASD 2023

作者： Jagan, S. Murugan, G. Rajasathiya, K. Badhoutiya, Arti Hemalatha, S. Ilyos, Kabulov Alzubaidi, Laith H. Indupalli, Tabitha Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology CSE India E.G.S. Pillay Engineering College Department of Computer Science and Business System Tamil Nadu Nagapattinam India P.S.R.R College of Engineering Department of Computer Science and Engineering Sivakasi626140 India GLA University Department of Electrical Engineering Mathura India Padmavathy Engineering College EEE Prince Shri Venkateshwara Chennai127 India National Research University Tashkent Institute of Irrigation and Agricultural Mechanization Engineers Tashkent Uzbekistan The Islamic University Najaf Iraq GRIET Telangana Hyderabad India

ISBN: (纸本)9798350336474

This research proposes a novel approach that integrates distinct Artificial Intelligence (AI) methodologies to address the issue of energy demand planning in Smart Homes (SH). The novel technique is designed to address a multi-objective scheduling issue, aiming to strike a balance between energy consumption and user convenience. The AI-based Power Scheduling issue in a Smart Home (AI-PSSH) involves efficiently scheduling intelligent home equipment within specified constraints and flexible pricing schemes. Various methods have been suggested to tackle the problem of AI-PSSH effectively, including both optimal and non-optimal strategies. This study presents a novel formulation for the Intelligent Home Batteries (IHB) in the context of AI-PSSH, which mitigates the impact of constraints in achieving the optimum or near-optimum outcomes. IHB may optimize the scheduling of intelligent home devices by saving excess power when it is not needed and using saved power when it is more suited to achieve the goals of PSSH. PSSH is a multi-objective optimization problem aimed at concurrently achieving all goals. Incorporating AI into smart homes has facilitated the development of sophisticated energy management techniques in scheduling and power optimization. The approach utilizes Particle Swarm Optimization (PSO) to improve the effectiveness of scheduling algorithms. Integrating PSO into the scheduling process enhances flexibility and enables smart decision-making based on real-time adjustments. The simulation results assess the proposed multi-objective AI-PSSH utilizing the IHB (MO-AI-PSSH) strategy by considering five possibilities of energy use and flexible pricing strategies. The efficiency of the suggested MO-AI-PSSH technique is compared to that of existing cutting-edge algorithms using the suggested databases utilizing the provided datasets. The MO-AI-PSSH technique, as suggested, demonstrates superior performance compared to the other algorithms in almost all instances. ©

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Novel orange-red glaze based on in-doped CaFe2O4 red chromophore: Synthesis, properties and characterizations

引用

Ceramics International 2025年

作者： Qi, Menghang Tian, Liangsheng Wang, Qiuying Xiao, Ruoxiu Hu, Nan Shi, Lin Siritanon, Theeranun Jiang, Peng Department of Inorganic Nonmetallic Materials School of Materials Science and Engineering University of Science and Technology Beijing Beijing100083 China School of Computer and Communication Engineering University of Science and Technology Beijing Beijing100083 China Institute of Surface-Earth System Science School of Earth System Science Tianjin University Tianjin300072 China School of Chemistry Institute of Science Suranaree University of Technology 111 University Avenue Muang Nakhon Ratchasima30000 Thailand

The CaFe2-xInxO4 (0 ≤ x ≤ 1) pigment with red shade was successfully synthesized through solid state reactions. The crystal structure, color development mechanism, morphology and its application in the glaze were studied. XPS analysis confirmed Fe resides in +3 oxidation state. Combined with UV–Vis–NIR analysis, it is determined that the formation of red shade of the pigment is due to the 6A1(S) → 4T2(D), 6A1(S) → 4A1, 4E(G) and 6A1(S) → 4T2(G) transitions of Fe3+ in the visible light region. The pigment was applied to the lead-free low temperature glaze to evaluate its potential in preparing new glaze color. The as-synthesized glaze showed orange-red shade. With increased In3+ doping concentration, the glaze demonstrates a stronger red tone. The glaze also shows high NIR solar reflectance. Inferred from SEM-EDS, the pigment particles are evenly scattered in the glaze. Therefore, CaFe2-xInxO4 (0 ≤ x ≤ 1) red pigment is a qualified candidate for red shade glaze preparation. © 2025 Elsevier Ltd and Techna Group S.r.l.

关键词： Crystal structure

来源：评论

学校读者我要写书评

暂无评论

A Trajectory-Based Stochastic Approach to Symbolic Control

A Trajectory-Based Stochastic Approach to Symbolic Control

引用

IEEE Conference on Decision and Control

作者： Alessandro Tenaglia Corrado Possieri Daniele Carnevale Department of Civil Engineering and Computer Science Engineering University of Rome Tor Vergata Rome Italy National Research Council of Italy Institute for System Analysis and Computer Science “A. Ruberti” Rome Italy

This paper presents two innovative approaches to design symbolic controllers for dynamical systems. The first novelty involves a new trajectory-based strategy for defining the states of a symbolic model, which provides a more accurate representation of the system's dynamics than the traditional grid-based technique. The second novelty concerns using a Bounded-parameter Markov Decision Process rather than a Finite Transition system to model the behavior of a symbolic model. This procedure allows for handling the system's stochastic behavior and considers uncertainties. The effectiveness of the novel approaches presented is demonstrated through numerical results.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Trajectory Control and Fair Communications for Multi-UAV Networks: A Federated Multi-Agent Deep Reinforcement Learning Approach

引用

IEEE Transactions on Wireless Communications 2025年

作者： Tarekegn, Getaneh Berie Tesfaw, Belayneh Abebe Juang, Rong-Terng Saha, Dola Tarekegn, Robel Berie Lin, Hsin-Piao Tai, Li-Chia Nycu Department of Electrical and Computer Engineering 1001 University Road Hsinchu300 Taiwan Nycu Institute of Electrical and Control Engineering 1001 University Road Hsinchu300 Taiwan National Taipei University of Technology Department of Electrical Engineering and Computer Science Taipei10608 Taiwan National Taipei University of Technology Institute of Aerospace and System Engineering Taipei10608 Taiwan University at Albany Suny Department of Electrical and Computer Engineering AlbanyNY12222 United States

Due to their highly flexible deployment and agility features, unmanned aerial vehicles (UAVs) serving as aerial base stations are increasingly being used in challenging environments, including emergency communication, traffic offloading, and failures of existing communications infrastructure. A reliable and effective communication service requires 3D deployment and autonomous UAV trajectory optimization in each time slot. While most prior approaches focus on trajectory design that maximizes communication coverage or network throughput without considering fairness as well as computation time. This paper presents a multi-UAV trajectory control and fair communication (TCFC) scheme that maximizes ground user data rate and communication coverage while ensuring service fairness in a UAV-aided wireless communications system. The proposed TCFC scheme employs a two-stage learning approach. Firstly, it develops a gated recurrent unit-based link quality estimation model to assess each user's link quality over time. Then, a federated multi-agent deep reinforcement learning (FedMADRL) algorithm is utilized to continuously adjust the trajectory of the UAVs, optimizing communication performance. We evaluated our proposed system using real channel measurement data, i.e., amplitude and phase signal information. The results show that the proposed TCFC scheme reduces computation time by 26.91% and provides comparable network performance with the baseline methods while improving DRL agents' privacy. © 2025 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent Deep Reinforcement Learning for Access Point Activation Strategy in Cell-Free Massive MIMO Networks

Multi-Agent Deep Reinforcement Learning for Access Point Act...

引用

IEEE Conference on computer Communications Workshops, INFOCOM Wksps

作者： Li Sun Jing Hou Richard Chapman Department of Computer Science and Software Engineering Auburn University Auburn AL USA Department of Computer Science and Information System California State University San Marcos San Marcos CA USA

Cell-free massive MIMO network is a promising solution to support 5G communication. However, as an ultra-dense network (UDN), it suffers from two major issues: power consumption and scalability, which hinders it to be widely used in practice. The increasing and time-varying user demand, the dynamic propagation environment, and the huge amount of access points (APs) make it a challenging task to address these issues. For this purpose, we propose a distributed solution to solve the access point activation (APA) problem in cell-free massive MIMO networks to reduce power consumption considering dynamic user demand. We leverage the user-centric characteristic and design a multi-agent deep reinforcement learning (MADRL) algorithm by which each AP independently decides whether it needs to be switched on or off. The simulation results show that the MADRL outperforms the centralized and the random strategies. This research demonstrates the ability of distributed learning in solving the APA problem in cell-free massive MIMO networks.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Investigation and Development into the Use of AI-Based Analytical Methods for Forecasting the Stock Market 4

An Investigation and Development into the Use of AI-Based An...

引用

4th International Conference on Smart Electronics and Communication, ICOSEC 2023

作者： Sheeba, M. Kumar, D. Ranjith Kirubakaran, S. Dixit, Ila Rani, D. Leela Yadav, Ajay Singh Department of B. Com Information System Management Tamil Nadu Chennai600042 India Department of B. Com Corporate Secretaryship Tamil Nadu Chennai600042 India CMR College of Engineering and Technology Department of Computer Science and Engineering Telangana Hyderabad501401 India Maharaja Agarsen International College Department of Management Chhattisgarh Raipur492001 India School of Engineering Mohan Babu University Department of Electronics and Communication Engineering Andhra Pradesh Tirupati517102 India Srm Institute of Science and Technology Department of Mathematics Utter Pradesh Ghaziabad India

ISBN: (纸本)9798350300888

Forecasting the stock market is difficult due to the fact that a number numerous factors, including the current situation of economy, current politics, and market sentiment. Traditional analytical methods have limitations in accurately predicting stock prices, leading to increased curiosity about using algorithms based on artificial intelligence (AI) and machine learning (ML) stock market forecasting. This paper investigates the use of AI-based analytical methods for forecasting the stock market and presents a detailed analysis of various algorithms used in AI-based stock market forecasting. The use of AI-based analytical methods for forecasting the stock market has several benefits. Firstly, AI-based models can analyse vast amounts of financial data much faster than traditional methods, allowing for more accurate and timely predictions. Secondly, AI-based models can handle complex relationships between different factors affecting stock prices, such as market trends, news events, and economic indicators, which can be challenging to capture using traditional methods. This study also discusses about the challenges and limitations of using AI-based models for stock market forecasting research in the future research. Here, this study uses a novel theory of recurrent neural networks and a random forest model to enable prediction with more accuracy. One of the commonly used algorithms in AI-based stock market forecasting is the recurrent neural network (RNN), which model that uses deep learning. RNNs are very helpful for modelling time-series information, like stock prices, just because they may ensnare the temporal data dependencies. In addition, RNNs can be used to generate forecasts by predicting the next value in a set of timestamps based on previous values. Another popular algorithm for stock market forecasting is the Random Forest (RF) ensemble model. RF models are a type of decision tree-based algorithm that can generate accurate predictions by combining the output

关键词： Financial markets

来源：评论

学校读者我要写书评

暂无评论

Current Problems and Proposed Solutions for the Electronic Commerce Sector 1

Current Problems and Proposed Solutions for the Electronic C...

引用

1st International Conference on computer science and Emerging Technologies, CSET 2023

作者： Wahshat, Hassan Al Al-Rousan, Riyad Al-Haithami, Wafa Abdulkarem Ahmad Saany, Syarilla Iryani Mohamad, Fatma Susilawati Kumar, Rohit Ajloun National University Faculty of Business Department of Management Information System Ajloun Jordan Department of Virtual and Augmented Reality Jordan Mediu Faculty of Finance and Administrative Sciences Malaysia UniSZA Faculty of Informatics and Computing Malaysia School of Computer Science and Engineering Jain Deemed to Be University Bangalore India

ISBN: (纸本)9798350341737

The rapid growth of electronic commerce (e-commerce) has revolutionized the global business landscape, transforming the way consumers and businesses interact, purchase, and sell goods and services. However, this sector faces a myriad of challenges that impede its sustainable growth and development. This research paper critically examines the current problems encountered by the electronic commerce sector and proposes viable solutions to address these issues. The study identifies three major problem areas affecting the sector: (1) Security and Privacy Concerns, (2) Logistics and Supply Chain Bottlenecks, and (3) Customer Experience and Trust. Regarding security and privacy, the proliferation of cyber-attacks, data breaches, and identity theft has eroded consumer confidence, hindering e-commerce adoption. The paper aims to investigate the latest technologies and best practices, such as blockchain, advanced encryption techniques, and robust authentication methods, to fortify e-commerce platforms against cyber threats and safeguard user data. The second problem area, logistics and supply chain bottleneck, poses challenges related to order fulfilment, inventory management, and last-mile delivery. Through an in-depth analysis, the research proposes innovative approaches like utilizing machine learning algorithms for predictive inventory management, integrating Internet of Things (IoT) devices for real-time tracking, and exploring drone and autonomous vehicle deliveries to enhance overall logistics efficiency. The third issue addressed in this paper pertains to customer experience and trust. The inability to tangibly experience products before purchase, coupled with ambiguous return policies, often deters potential customers from engaging in e-commerce transactions. The study delves into augmented reality (AR) and virtual reality (VR) applications to simulate product experiences and advocates for transparent and customer centric return policies. This research paper sheds li

关键词： Supply chains

来源：评论

学校读者我要写书评

暂无评论

Graphed-based K-Means and Shortest Distance Tree for the Construction of Elderly Safe Corridor Accident and Prevention Platform

Graphed-based K-Means and Shortest Distance Tree for the Con...

引用

International Conference on Advanced Communication Technology (ICACT)

作者： Ying-Jhen Sua Zhen-Kai Wua You-Jia Sub Yu-Shan Linb Yao-Chung Changa Department of Computer Science and Information Engineering National Taitung University Taitung Taiwan Department of Information Science and Management System National Taitung University Taitung Taiwan

As medicine, science, and technology improve, the average age of the population on earth is also increasing. The World Health Organization (WHO) reported that the average age of the countries that had joined and surveyed was 68 years old, and the highest country had over 80 years. The care of the elderly population requires more attention and careful planning. There are a variety of social support systems available for the elderly today. Apart from encouraging them to participate in various social affairs and activities, it is also necessary to consider the safety and accident prevention needs of the elderly when they move. This is one of the critical factors for the elderly to obtain various opportunities and resources. Those areas or locations where older adults are densely concentrated require a more effective police presence to maintain the safety of moving routes. This research primarily deals with arranging police forces based on a simulation of elderly individuals' action trajectories. The level of care that the elderly require is weighted. Furthermore, the weighted K-means Clustering method is used to find the clusters of the elderly and make the cluster center close to the elderly who need high care to deploy the police force to provide higher priority services. Then the shortest paths between the elderly settlements (cluster center) and the activity venue are found based on the consideration that the movement patterns of the elderly are mainly walking, and they are most likely to arrive at the activity venue within the shortest distance. The activity venues of the shortest paths are taken as the root to construct the shortest path tree. During tree construction, the number of elderly in each settlement is summed in each intersection node as the number of elderly passing through. Then the breadth-first search method is applied to find the multi-path convergence points that the most elderly pass through as the reference for recommending police force arrangem

关键词： Law enforcement Search methods Force Sociology Data models Safety Trajectory

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：