检索结果-内蒙古大学图书馆

Open Innovations Association (FRUCT), Conference of

作者： Aisha Alabsi Ammar Hawbani Xingfu Wang Saeed Hamood Alsamhi Liang Zhao Ahmed Al-Dubai University of Science and Technology of China School of Computer Science and Technology Insight Centre for Data Analytics National University of Ireland Galway Ireland Shenyang Aerospace University School of Computer Science Computing school of Edinburgh Napier University United Kingdom

With the rise of computing-intensive applications like online gaming and telemedicine on user equipment (UE) and the evolution of 5G technology, there is a surge in demand for greater computing resources and power. Yet, UEs have limited resources and batteries. Mobile Cloud Computing (MCC) has emerged as a method to enhance UEs computing capabilities and conserve energy by transferring tasks to the cloud. Mobile Edge Computing (MEC) further aids by reducing delays, although it faces issues like limited resources and unpredictable network conditions. Unmanned Aerial Vehicles (UAVs) offer a remedy by serving as mobile stations for MEC, but optimal offloading decisions in UAV-assisted MEC remain intricate. Addressing this, I propose using Reinforcement Learning (RL), specifically Q-Learning, Deep Q Network (DQN), and Deep Deterministic Policy Gradient (DDPG), to enhance decision-making for offloading. Our focus is on energy efficiency and reduced service delay, and our simulations prove our method's efficacy in UAV-assisted MEC environments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Chroma Intra Prediction via Exemplar Colorization Network for Versatile Video Coding

引用

IEEE Transactions on Multimedia 2025年

作者： Pan, Zhaoqing Chen, Jixing Peng, Bo Lei, Jianjun Wang, Fu Lee Ling, Nam Kwong, Sam Tianjin University School of Electrical and Information Engineering Tianjin300072 China Hong Kong Metropolitan University School of Science and Technology Hong Kong Hong Kong Santa Clara University Department of Computer Science and Engineering Santa ClaraCA95053 United States Lingnan University School of Data Science Hong Kong Hong Kong

Chroma intra prediction aims to reduce chroma redundancies within a frame, which plays an important role in improving the coding efficiency of intra coding. Existing chroma intra prediction methods typically utilize the spatial relationship between the current luma block and its neighboring reference luma blocks to predict its chroma samples. However, the spatial properties of luma components differ from those of chroma components, which limits the accuracy of chroma intra prediction. To tackle this issue, an efficient Exemplar Colorization Network (ECNet)-based chroma intra prediction method is proposed in this paper, in which the colorization relationship between reference luma and chroma components is exploited to predict the chroma components for the current luma component. Inspired by the principle that semantic information in an image exhibits short-range continuity, a Spatial-consistency-based Colorization Transfer Network (SCTNet) is proposed, which builds and transfers colorization representations of neighboring reference blocks for chroma prediction. To improve the chroma prediction capability of SCTNet, a colorization learning module is developed to learn the robust mapping relationship from the luma component to the chroma component in a region-to-pixel manner, and a weight-adaptive reconstruction module is designed to adaptively utilize reference information from neighboring blocks to generate an initial prediction result. In addition, to further improve the accuracy of chroma intra prediction, a multi-reference-based chroma refinement network is proposed, which simultaneously uses the spatial information of neighboring reference chroma blocks and the current luma block to eliminate blocking and color-bleeding artifacts in the initial prediction result. Experimental results demonstrate that our proposed ECNet outperforms the state-of-the-art chroma intra prediction methods in terms of coding performance. © 1999-2012 IEEE.

关键词： Video signal processing

来源：评论

学校读者我要写书评

暂无评论

Exploring and Enhancing the Transfer of Distribution in Knowledge Distillation for Autoregressive Language Models

arXiv

引用

arXiv 2024年

作者： Rao, Jun Liu, Xuebo Lin, Zepeng Ding, Liang Li, Jing Tao, Dacheng Zhang, Min The Harbin Institute of Technology Shenzhen518000 China The School of Computer Science University of Sydney NSW2006 Australia The College of Computing & Data Science Nanyang Technological University Singapore

Knowledge distillation (KD) is a technique that compresses large teacher models by training smaller student models to mimic them. The success of KD in auto-regressive language models mainly relies on Reverse KL for mode-seeking and student-generated output (SGO) to combat exposure bias. Our theoretical analyses and experimental validation reveal that while Reverse KL effectively mimics certain features of the teacher distribution, it fails to capture most of its behaviors. Conversely, SGO incurs higher computational costs and presents challenges in optimization, particularly when the student model is significantly smaller than the teacher model. These constraints are primarily due to the immutable distribution of the teacher model, which fails to adjust adaptively to models of varying sizes. We introduce Online Knowledge Distillation (OKD), where the teacher network integrates small online modules to concurrently train with the student model. This strategy abolishes the necessity for on-policy sampling and merely requires minimal updates to the parameters of the teacher’s online module during training, thereby allowing dynamic adaptation to the student’s distribution to make distillation better. Extensive results across multiple generation datasets show that OKD achieves or exceeds the performance of leading methods in various model architectures and sizes, reducing training time by up to fourfold. Copyright © 2024, The Authors. All rights reserved.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Hotel Information System Management Audit Using COBIT 2019 - APO12

Hotel Information System Management Audit Using COBIT 2019 -...

引用

Innovation in Technology (INOCON), IEEE International Conference for

作者： Joy Nashar Utamajaya Suhono Harso Supangat Ford Lumban Gaol Benny Ranti BINUS Graduate Program School of Information System BINUS University Jakarta Indonesia School of Electrical Engineering and Informatic Bandung Institute of Technology Bandung Indonesia Computer Science Department BINUS Graduate Program Computer Science Bina Nusantara University Jakarta Indonesia Computer Science Department Faculty of Computer Science University of Indonesia Depok Indonesia

In general, public or private organizations or companies have used information-based technology as a support to improve business performance to be more effective and efficient in order to achieve a company’s business goals. Given the large contribution of information technology in the application of hotel applications as a supporting information system, it is a vital system that must avoid risks that can hinder and cause harm to the hotel management business processes. Therefore, it is necessary to carry out risk management using the COBIT 5 framework to manage possible risks that may occur based on the APO12 (Manage Risk) domain. From the evaluation results of the data obtained through observation and interviews as well as the calculation of the results of the questionnaire based on 6 APO12 subdomains, the results of the risk management level capability assessment in hotel applications are still at level 3 or have reached the level of established process with the target to be achieved at level 4 resulting in a gap of 1 level. Based on the results obtained, it is necessary to propose recommendations that can be used by hotels in improving the application of information technology risk management so that in the future it can achieve the expected target level so that the APO12 level of capability can increase and become more optimal.

关键词： Technological innovation Companies Risk management Information technology Interviews Information systems

来源：评论

学校读者我要写书评

暂无评论

M2ConceptBase: A Fine-Grained Aligned Concept-Centric Multimodal Knowledge Base

arXiv

引用

arXiv 2023年

作者： Zha, Zhiwei Wang, Jiaan Li, Zhixu Zhu, Xiangru Song, Wei Xiao, Yanghua Shanghai Key Laboratory of Data Science School of Computer Science Fudan University Shanghai China School of Computer Science and Technology Soochow University Suzhou China Research Center for Intelligent Robotics Zhejiang Lab Hangzhou China

Multimodal knowledge bases (MMKBs) provide cross-modal aligned knowledge crucial for multimodal tasks. However, the images in existing MMKBs are generally collected for entities in encyclopedia knowledge graphs. Therefore, detailed groundings of visual semantics with linguistic concepts are lacking, which are essential for the visual concept cognition ability of multimodal models. Addressing this gap, we introduce M2ConceptBase, the first concept-centric MMKB. M2ConceptBase models concepts as nodes with associated images and detailed textual descriptions. We propose a context-aware multimodal symbol grounding approach to align concept-image and concept-description pairs using context information from image-text datasets. Comprising 951K images and 152K concepts, M2ConceptBase links each concept to an average of 6.27 images and a single description, ensuring comprehensive visual and textual semantics. Human studies confirm more than 95% alignment accuracy, underscoring its quality. Additionally, our experiments demonstrate that M2ConceptBase significantly enhances VQA model performance on the OK-VQA task. M2ConceptBase also substantially improves the fine-grained concept understanding capabilities of multimodal large language models through retrieval augmentation in two concept-related tasks, highlighting its value.1 Copyright © 2023, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

MMFNet: A Multi-modal and Multi-attention Fusion Network for Multi-task Skin Disease Classification

MMFNet: A Multi-modal and Multi-attention Fusion Network for...

引用

2024 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2024

作者： Zhu, Xinlei Qian, Pengjiang Wang, Chuang Liu, Li Gao, Ming Cai, Weiwei Ng, Eddie Yin-Kwee Jiangnan University School of Artificial Intelligence and Computer Science Wuxi China Jiangnan University Engineering Research Center of Intelligent Technology for Healthcare Ministry of Education Wuxi China Affiliated Hospital Jiangnan University Big Data Center Wuxi China Nanyang Technological University School of Mechanical and Aerospace Engineering Nanyang Singapore

ISBN: (纸本)9798350386226

Skin diseases rank among the most prevalent ailments in humans, which underscores the critical importance of early detection and diagnosis. Considering clinical practice, the integration of information from various modalities holds significant potential to enhance diagnostic accuracy and precision. In this study, we propose a multi-modal and multi-attention fusion network for multi-task skin disease classification. We initially devise the Shifted Window Cross-attention Fusion (SWCF) module based on the shifted window self-attention mechanism, leveraging the advantages of the attention mechanism to learn the correlations between different modalities and fully integrate multi-modal features. Subsequently, taking into account the heterogeneity of the data, we employ the Heterogeneous data Cross-attention Fusion (HDCF) module using the idea of information decoupling and separation processing to integrate image and text data. Additionally, we suggest the Dynamic Loss Weight Allocation (DLWA) method for multi-task learning to refine the training procedure. We confirm the superiority of the proposed method on the publicly available multi-modal skin lesion dataset, Derm7pt. The average accuracy of multi-modal skin disease classification is 79.14%, surpassing the current state-of-the-art methods. © 2024 IEEE.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Process Mining to Discover and Preserve Infrequent Relations in Event Logs: An Application to Understand the Laboratory Test Ordering Process Using the MIMIC-III dataset 32

Process Mining to Discover and Preserve Infrequent Relations...

引用

32nd Australasian Conference on Information Systems: Information Systems for a Sustainable Future, Connectedness, and Social Good, ACIS 2021

作者： Chen, Qifan Lu, Yang Tam, Charmaine S. Poon, Simon K. School of Computer Science The University of Sydney Sydney Australia Centre for Translational Data Science Northern Clinical School The University of Sydney Sydney Australia

Process mining techniques can provide insights into the healthcare domain with the rapid growth of electrical health records. Process mining is about understanding the sequence of activities in event logs, where directly-follows relations identify pairs of activities that follow each other directly. Existing research explores frequent relations, while infrequent relations are often seen as noises and filtered out during discovery. However, important insights may be revealed through these infrequent relations, especially in healthcare processes. This paper aims to use process mining techniques to discover and preserve value-based conditional infrequent relations. We adopt the L* life-cycle methodology and data-aware Heuristic Miner (DHM) as tools to provide a worded example based on extracted data from the MIMIC-III dataset, which is a publicly available database containing a large amount of electrical health records (EHR), to show how process mining can be used to analyse infrequent relations in a laboratory test’s ordering process. Copyright © 2021 Chen, Lu, Tam, & Poon.

关键词： Statistical tests

来源：评论

学校读者我要写书评

暂无评论

Trace-based Multi-Dimensional Root Cause Localization of Performance Issues in Microservice Systems

Trace-based Multi-Dimensional Root Cause Localization of Per...

引用

International Conference on Software Engineering (ICSE)

作者： Chenxi Zhang Zhen Dong Xin Peng Bicheng Zhang Miao Chen Fudan University China Shanghai Collaborative Innovation Center of Intelligent Visual Computing China School of Computer Science and Shanghai Key Laboratory of Data Science Fudan University China

ISBN: (数字)9798400702174

ISBN: (纸本)9798350382143

Modern microservice systems have become increasingly complicated due to the dynamic and complex interactions and runtime environment. It leads to the system vulnerable to performance issues caused by a variety of reasons, such as the runtime environments, communications, coordinations, or implementations of services. Traces record the detailed execution process of a request through the system and have been widely used in performance issues diagnosis in microservice systems. By identifying the execution processes and attribute value combinations that are common in anomalous traces but rare in normal traces, engineers may localize the root cause of a performance issue into a smaller scope. However, due to the complex structure of traces and the large number of attribute combinations, it is challenging to find the root cause from the huge search space. In this paper, we propose TraceContrast, a trace-based multidimensional root cause localization approach. TraceContrast uses a sequence representation to describe the complex structure of a trace with attributes of each span. Based on the representation, it combines contrast sequential pattern mining and spectrum analysis to localize multidimensional root causes efficiently. Experimental studies on a widely used microservice benchmark show that TraceContrast outperforms existing approaches in both multidimensional and instance-dimensional root cause localization with significant accuracy advantages. Moreover, Trace-Contrast is efficient and its efficiency can be further improved by parallel execution.

关键词： Location awareness Runtime environment Accuracy Microservice architectures Benchmark testing data mining Spectral analysis

来源：评论

学校读者我要写书评

暂无评论

Ghost-in-Wave: How Speaker-Irrelative Features Interfere DeepFake Voice Detectors

Ghost-in-Wave: How Speaker-Irrelative Features Interfere Dee...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Xuan Hai Xin Liu Zhaorun Chen Yuan Tan Song Li Weina Niu Gang Liu Rui Zhou Qingguo Zhou School of Information Science and Engineering Lanzhou University Lanzhou China Elmore Family School of Electrical and Computer Engineering Purdue University IN USA The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Recent speech synthesis technology can generate high-quality speech indistinguishable from human speech, thus introducing various security and privacy risks. Numerous recent studies have focused on fake voice detection to address these risks, with many claiming to achieve ideal performance. However, is this really the case? A recent research work introduced Speaker-Irrelative-Features (SiFs), unrelated to the information in speech files but capable of influencing fake detectors. This means that existing detectors may rely on SiFs to a certain extent to distinguish real and fake speech. In this paper, we introduce an evaluation framework to evaluate the influence of SiFs in existing fake voice detectors in depth. We evaluate three SiFs which include background noise, the mute parts before and after voice, and the sampling rate on ASVspoof2019 and FoR. Our results confirm the substantial influence of SiFs on fake voice detection performance, and we delve into the analysis of the underlying mechanisms.

关键词： Privacy Deepfakes Detectors Feature extraction Background noise Speech synthesis Security

来源：评论

学校读者我要写书评

暂无评论

Rethinking Local and Global Feature Representation for Semantic Segmentation 32

Rethinking Local and Global Feature Representation for Seman...

引用

32nd British Machine Vision Conference, BMVC 2021

作者： Chen, Mohan Zhao, Xinxuan Fu, Bingfei Zhang, Li Xue, Xiangyang Academy for Engineering & Technology Fudan University China School of Data Science Fudan University China School of Computer Science Fudan University China

Although fully convolution networks (FCN) have dominated semantic segmentation since the birth of [24], they are inherently limited in capturing long-range structured relationship with the layers of local kernels. While recent Transformer-based models have proven extremely successful in computer vision tasks by capturing global representation, they would deteriorate semantic segmentation by over-smoothing the regions contain fine details (e.g., boundaries and small objects). To this end, we propose a Dual-Stream Convolution-Transformer segmentation framework, called DSCT, by taking advantage of both the convolution and Transformer to learn a rich feature representation for semantic segmentation. Specifically, DSCT extracts high resolution local feature information from convolution layers and global feature representation across the Transformer layers. Moreover, a feature fusion module is plugged to exchange information between spatial stream and context stream at each stage. With the local and global context modeled explicitly in every layer, the two streams can be combined with a simple decoder to provide a powerful segmentation model. Extensive experiments show that our model builds a new state of the art on Cityscapes dataset (83.31% mIoU) with only 80K training iterations and appealing performance (49.27% mIoU) on ADE20K, outperforming most of the alternatives with a new perspective. © 2021. The copyright of this document resides with its authors.

关键词： Convolution

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：