检索结果-内蒙古大学图书馆

Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Nunung Nurul Qomariyah Ahmad Nurul Fajar Vincent Yono Dimitar Kazakov Computer Science Department School of Computing and Creative Arts Bina Nusantara University Jakarta Indonesia Information Systems Management Department Binus Graduate Program Master of Information System Management Bina Nusantara University Jakarta Indonesia Computer Science Department University of York United Kingdom

This paper explores the development of a multilabel machine learning system for predicting both gender and age from human gait patterns. Gait analysis, a non-intrusive method of identifying subtle nuances in human movement, has proven to be a rich source of information related to demographic characteristics. The research extends beyond traditional single-label classification approaches, adopting a multilabel framework to simultaneously predict gender and age *** study evaluates various multilabel machine learning algorithms, with the Random k-labeLsets (RAKEL) algorithm demonstrating superior performance in predicting gender and age labels from human gait datasets. The accuracy of the algorithm can reach up to 87%. We also compared the multilabel approach with several multiclass algorithms such as Decision Tree, Random Forest, Gradient Boosting, K-Neighbors and XGBoost. However, when we also considering the training time, Classifier Chain algorithm showed the best trade off with the accuracy of 86% and the training time is twice faster than the RAKEL algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Adaptive Contextual Caching for Mobile Edge Large Language Model Service

arXiv

引用

arXiv 2025年

作者： Liu, Guangyuan Liu, Yinqiu Wang, Jiacheng Du, Hongyang Niyato, Dusit Kang, Jiawen Xiong, Zehui College of Computing and Data Science The Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore College of Computing and Data Science Nanyang Technological University Singapore Department of Electrical and Electronic Engineering University of Hong Kong Hong Kong School of Automation Guangdong University of Technology China Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore

Mobile edge Large Language Model (LLM) deployments face inherent constraints, such as limited computational resources and network bandwidth. Although Retrieval-Augmented Generation (RAG) mitigates some challenges by integrating external knowledge bases, inefficient cache management can still result in high retrieval latency and frequent cache updates. To address these issues, we propose an Adaptive Contextual Caching (ACC) framework that anticipates user needs by proactively caching semantically relevant data for mobile-edge LLMs. ACC utilizes a deep reinforcement learning (DRL) module to refine cache replacement policies, balancing user context, document similarity, and the overhead associated with cache misses. Experimental results demonstrate that ACC increases cache hit rates to over 80% after only 11 training episodes, outperforming FIFO, LRU, and semantic-only caching while reducing retrieval latency by up to 40%. In particular, ACC also reduces local caching overhead (i.e., the cost of updating the cache when a miss occurs) by as much as 55%, enabling scalable, low-latency LLM services in resource-constrained edge environments. © 2025, CC BY.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

ETHICS-2023 Session D1 - Open Forum: Building a Technology Ethics Community

ETHICS-2023 Session D1 - Open Forum: Building a Technology E...

引用

IEEE International Symposium on Ethics in Science, Technology and Engineering

作者： Thomas Creely Marc Cheong Heather Love Ketra Schmitt Ethics and Emerging Military Technology Graduate Program U.S. Naval War College School of Computing and Information Systems & Centre for AI and Digital Ethics (CAIDE) University of Melbourne English Language and Literature University of Waterloo Concordia University

This session was an open forum where audience members were invited to participate in discussions of a number of themes with relevance to ethics and technology and to future conferences in the IEEE ETHICS series. The discussions took place in small groups, with groups reporting back to the full cohort for collaborative brainstorming.

关键词：

来源：评论

学校读者我要写书评

暂无评论

2D Basement Relief Inversion using Sparse Regularization.

arXiv

引用

arXiv 2024年

作者： Barboza, Francisco Márcio da Cunha Romão Silva, E. Arthur Anthony de Carvalho, Bruno Motta Department of Computing and Technology Federal University of Rio Grande do Norte Rio Grande do Norte Caicó Brazil Graduate Program on Systems and Computing Federal University of Rio Grande do Norte Rio Grande do Norte Natal Brazil Department of Informatics and Applied Mathematics Federal University of Rio Grande do Norte Rio Grande do Norte Natal Brazil

Basement relief gravimetry is a key application in geophysics, particularly important for oil exploration and mineral prospecting. It involves solving an inverse geophysical problem, where the parameters of a geological model are inferred from observed data. In this context, the geological model consists of the depths of constant-density prisms representing the basement relief, and the data correspond to the gravitational anomalies caused by these prisms. Inverse geophysical problems are typically ill-posed, as defined by Hadamard, meaning that small perturbations in the data can result in large variations in the solutions. To address this instability, regularization techniques, such as those proposed by Tikhonov, are employed to stabilize the solutions. This study presents a comparative analysis of various regularization techniques applied to the gravimetric inversion problem, including Smoothness Constraints, Total Variation, the Discrete Cosine Transform (DCT), and the Discrete Wavelet Transform (DWT) using Daubechies D4 wavelets. Optimization methods are commonly used in inverse geophysical problems because of their ability to find optimal parameters that minimize the objective function—in this case, the depths of the prisms that best explain the observed gravitational anomalies. The Genetic Algorithm (GA) was selected as the optimization technique. GA is based on Darwinian evolutionary theory, specifically the principle of natural selection, where the fittest individuals in a population are selected to pass on their traits. In optimization, this translates to selecting solutions that most effectively minimize the objective function. The results, evaluated using fit metrics and cumulative error analysis, demonstrate the effectiveness of all the regularization techniques and the Genetic Algorithm. Among the methods tested, the Smoothness constraint was briefly the most effective for the first and second synthetic models. For the third model, which was based on re

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

TEAM: Task-Clustering and Enhanced Adaptability for Meta-Reinforcement Learning on Robotics Through Multi-Task Diffusion and Optimization

TEAM: Task-Clustering and Enhanced Adaptability for Meta-Rei...

引用

IEEE International Conference on Robotic computing (IRC)

作者： Joshua Ho Chien-Min Wang Chung-Ta King Chi-Wei Feng Shih-Yun Sunkuan Social Networks and Human-Centered Computing Program Taiwan International Graduate Program Institute of Information Science Academia Sinica Taipei Taiwan Institute of Information Systems and Applications National Tsing Hua University Hsinchu Taiwan Department of Computer Science National Tsing Hua University Hsinchu Taiwan Department of Electrical Engineering National Taiwan University Taipei Taiwan

ISBN: (数字)9798331521554

ISBN: (纸本)9798331521561

The ability of robots to imitate human learning strategies-rapidly adapting to new tasks without large datasets-has garnered significant attention in meta-learning. Meta-reinforcement learning seeks to enhance robotic agent flexibility across diverse tasks and contexts, offering promise where single-task learning often fails. Despite advancements like multi-task diffusion models and task-weighted optimization mechanisms, effectively training tasks with varying complexities simultaneously remains a major challenge. This paper introduces a novel meta-reinforcement learning method that addresses this issue by clustering the training tasks of robotic arms based on semantic and trajectory similarities, while leveraging adaptive learning rates and task-specific weights proposed by the multitask optimization techniques. Our approach, TEAM, emphasizes performance-driven semantic clustering, optimizing based on robotic task similarity, complexity, and convergence objectives. We also integrate fast adaptive and multi-task optimization of the diffusion model to enhance computational efficiency and adaptability. More specifically, we introduce a cluster-specific optimization technique, using specialized parameters for each group to allow more refined task handling. The experimental validation demonstrates the effectiveness of this scalable method in improving performance, adaptability, and efficiency in real-world, heterogeneous robotic tasks, further advancing robotic computing in meta-reinforcement learning.

关键词： Training Metalearning program processors Semantics Multitasking Diffusion models Complexity theory Trajectory Robots Optimization

来源：评论

学校读者我要写书评

暂无评论

How Good is ChatGPT at Audiovisual Deepfake Detection: A Comparative Study of ChatGPT, AI Models and Human Perception

arXiv

引用

arXiv 2024年

作者： Shahzad, Sahibzada Adil Hashmi, Ammarah Peng, Yan-Tsung Tsao, Yu Wang, Hsin-Min The Social Networks and Human-Centered Computing Program Taiwan International Graduate Program Academia Sinica Taipei Taiwan The Department of Computer Science National Chengchi University Taipei Taiwan The Institute of Information Systems and Applications National Tsing Hua University Hsinchu Taiwan The Research Center for Information Technology Innovation Academia Sinica Taipei Taiwan The Institute of Information Science Academia Sinica Taiwan

Multimodal deepfakes involving audiovisual manipulations are a growing threat because they are difficult to detect with the naked eye or using unimodal deep learning-based forgery detection methods. Audiovisual forensic models, while more capable than unimodal models, require large training datasets and are computationally expensive for training and inference. Furthermore, these models lack interpretability and often do not generalize well to unseen manipulations. In this study, we examine the detection capabilities of a large language model (LLM) (i.e., ChatGPT) to identify and account for any possible visual and auditory artifacts and manipulations in audiovisual deepfake content. Extensive experiments are conducted on videos from a benchmark multimodal deepfake dataset to evaluate the detection performance of ChatGPT and compare it with the detection capabilities of state-of-the-art multimodal forensic models and humans. Experimental results demonstrate the importance of domain knowledge and prompt engineering for video forgery detection tasks using LLMs. Unlike approaches based on end-to-end learning, ChatGPT can account for spatial and spatiotemporal artifacts and inconsistencies that may exist within or across modalities. Additionally, we discuss the limitations of ChatGPT for multimedia forensic tasks. © 2024, CC BY.

关键词： Audiovisual

来源：评论

学校读者我要写书评

暂无评论

AVTENet: A Human-Cognition-Inspired Audio-Visual Transformer-Based Ensemble Network for Video Deepfake Detection

引用

IEEE Transactions on Cognitive and Developmental systems 2025年

作者： Hashmi, Ammarah Shahzad, Sahibzada Adil Lin, Chia-Wen Tsao, Yu Wang, Hsin-Min Taiwan International Graduate Program Academia Sinica Social Networks and Human-Centered Computing Program Taipei11529 Taiwan National Tsing Hua University Institute of Information Systems and Applications Hsinchu30013 Taiwan National Chengchi University Department of Computer Science Taipei11605 Taiwan National Tsing Hua University Department of Electrical Engineering Institute of Communications Engineering Hsinchu300044 Taiwan Academia Sinica Research Center for Information Technology Innovation Taipei11529 Taiwan

The recent proliferation of hyper-realistic deepfake videos has drawn attention to the threat of audio and visual forgeries. Most previous studies on detecting artificial intelligence-generated fake videos only utilize visual modality or audio modality. While some methods exploit audio and visual modalities to detect forged videos, they have not been comprehensively evaluated on multimodal datasets of deepfake videos involving acoustic and visual manipulations, and are mostly based on convolutional neural networks with low detection accuracy. Considering that human cognition instinctively integrates multisensory information including audio and visual cues to perceive and interpret content and the success of transformer in various fields, this study introduces the audio-visual transformer-based ensemble network (AVTENet). This innovative framework tackles the complexities of deepfake technology by integrating both acoustic and visual manipulations to enhance the accuracy of video forgery detection. Specifically, the proposed model integrates several purely transformer-based variants that capture video, audio, and audio-visual salient cues to reach a consensus in prediction. For evaluation, we use the recently released benchmark multimodal audio-video FakeAVCeleb dataset. For a detailed analysis, we evaluate AVTENet, its variants, and several existing methods on multiple test sets of the FakeAVCeleb dataset. Experimental results show that the proposed model outperforms all existing methods and achieves state-of-the-art performance on Testset-I and Testset-II of the FakeAVCeleb dataset. We also compare AVTENet against humans in detecting video forgery. The results show that AVTENet significantly outperforms humans. © 2016 IEEE.

关键词： audio-visual AVTENet Deepfake detection

来源：评论

学校读者我要写书评

暂无评论

A Model-Driven Approach for Interoperability Among SaaS and DaaS/DBaaS: The MIDAS Case

引用

SN Computer Science 2022年第4期3卷 301页

作者： Mane, Babacar Mascarenhas, Ana Patrícia F. Magalhães Quinteiro, Gustavo Maciel, Rita Suzana Pitangueira Claro, Daniela Barreiro Formalisms and Semantic Applications Research Group (FORMAS) Institute of Computing Federal University of Bahia Av. Milton Santos Ondina Bahia Salvador 40170-110 Brazil Models for Software Engineering (MAtisse) Department of Exact Sciences and Earth State University of Bahia 2555 Silveira Martins St. Cabula Bahia Salvador Brazil Post Graduate Program in Computing and Systems Salvador University 251 Dr. Jose Peroba St. Bahia Salvador Brazil Models for Software Engineering (MAtisse) Institute of Computing Federal University of Bahia Av. Milton Santos Ondina Bahia Salvador 40170-110 Brazil

Cloud Platforms are heterogeneous, and users may face interoperability issues migrating applications or exchanging data among distinct clouds due, for instance, to the lack of standards solutions. Several solutions have been proposed to overcome lock-in situations, and middleware platforms are one of them. A semantic interoperability solution named middleware for Data as a Service (DaaS)/Database as a Service (DBaaS) and Software as a Service (SaaS)—MIDAS has been developed to overcome this lock-in issue. It is an intermediate communication layer to retrieve data from DaaS or DBaaS through a Structured Query Language (SQL) or a Not Only SQL (NoSQL) created at the SaaS level. MIDAS is a platform for software execution, but software development needs support for its entire life cycle. Therefore, we propose the Model drIven Approach for MIDAS (MIAMI), which enables the specification of platform-independent middleware models and their use to generate code on different cloud platforms. MIAMI comprises a Domain-Specific Modeling Language (DSML) that enables middleware models and a transformation specification, which defines how these models can be converted to code. MIAMI offers a strategy for MIDAS specification and code generation phases to help middleware developers’ activities. MIAMI was applied to code generation specifications in Cloud Foundry, Amazon Web Services, OpenShift, and Heroku providers to evaluate our approach. This specification shows MIAMI’s feasibility and points out that MDD is a promising approach to improving cloud interoperability solutions. © 2022, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Cloud computing Interoperability Metamodel Model-driven development

来源：评论

学校读者我要写书评

暂无评论

Classifying Ore Deposits Based on La-Icp-Ms Geochemistry of Natural Gold

SSRN

引用

SSRN 2024年

作者： Verbel, Angel Augusto Schutesky, Maria Emilia Gregory, Daniel D. Verbel, Arturo University of Brasilia Institute of Geosciences Graduate Program in Geology DF Brasília70910-900 Brazil University of Toronto Department of Earth Sciences TorontoONM5S 3B1 Canada Universidad de los Andes Systems and Computing Engineering Department School of Engineering Bogotá111711 Colombia

Systematic sampling of gold grains and their characterization in abundance and morphology can be used as a vectoring tool for mineral deposits. However, the processes that lead to the formation of an ore deposit are multiscale and nonlinear. As a result, the elemental compositions recorded in gold grains during the mineralization process seem mostly irregular and unpredictable. Here, we took advantage of the capabilities of neural networks to approximate, a natural nonlinear function that better explains the chemical variability observed in gold allowing us to make predictions. this was achieved by using published data of trace elements in gold of four types of gold deposits (i.e., Orogenic, VMS, porphyry and epithermal) from 47 different localities, obtained by laser ablation-inductively coupled plasma-mass spectrometry (LA- ICP-MS). The model was optimized by training five different architectures on the most influential elements, determined by principal component analysis (PCA). As a result, it was found that two hidden layers with 10 neurons each, is the best architecture for using trace elements as predictors of the type of deposit that formed a natural gold grain. In order to encourage the use of the findings made in this paper, we introduce OreGenes, an app developed in Matlab2023b based on the best-obtained model. It allows any user who possesses compatible data, the ability to import and process it to obtain a prediction with an average accuracy level of 88.9% confidence to determine the mineral deposit type that an unknown gold grain came from which grants the user the ability to have a powerful tool for exploration or research. © 2024, The Authors. All rights reserved.

关键词： Deposits

来源：评论

学校读者我要写书评

暂无评论

Mobile Apps for Healthy Life Education for Young People 2

Mobile Apps for Healthy Life Education for Young People

引用

2nd International Conference on Artificial Intelligence and Smart Energy, ICAIS 2022

作者： Suwandi, Benedictus Evan Warnars, Diana Teresia Spits Warnars, Harco Leslie Hendric Spits Utomo, Wiranto Herry Ramadhan, Arief Bina Nusantara University School of Computer Science Computer Science Department Jakarta11480 Indonesia Bina Nusantara University School of Information Systems Information System Department Jakarta11480 Indonesia Bina Nusantara University Binus Graduate Program - Computer Science Computer Science Department Jakarta11480 Indonesia Faculty of Computing President University Department of Magister Science of Information Technology Bekasi17550 Indonesia

ISBN: (纸本)9781665400527

Health education is crucial for the community, especially the younger generation, because adolescent behavior/lifestyle will carry over into adulthood, and it is difficult to change it. This paper proposes designing and implementing a mobile application for young people for healthy lifestyle education. This mobile application implemented with the android studio is designed using use case diagrams to describe the business processes of this mobile application. In addition, activity diagrams describe the process details of each activity use case, and class diagrams describe the design of the database model, which is implemented using a MySQL database. © 2022 IEEE.

关键词： Mobile computing

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：