检索结果-内蒙古大学图书馆

Deep Neural Network and GAN-Based Reversible Data Hiding in Encrypted Images: A Privacy-Preserving Approach

SN computer science 2024年第1期5卷 45页

作者： Nalavade, Jagannath E. Patil, Akshta Buchade, Amar Jadhav, Nagesh Department of Computer Science and Engineering MIT Art Design and Technology University School of Computing Pune India Department of Computer Engineering Rajarambapu Institute of Technology Sakharale Sangli India

In light of recent incidents involving the leakage of private photographs of Hollywood celebrities from iCloud, the need for robust methods to safeguard image content has gained paramount importance. This paper addresses this concern by introducing a novel framework for reversible image editing (RIT) supported by reversible data hiding with encrypted images (RDH-EI) techniques. Unlike traditional approaches vulnerable to hacking, this framework ensures both efficient and secure data embedding while maintaining the original image’s privacy. The framework leverages two established methods: secret writing and knowledge activity. While secret writing is susceptible to hacking due to the complex nature of cipher languages, RDH-EI-supported RIT adopts a more secure approach. It replaces the linguistic content of the original image with the semantics of a different image, rendering the encrypted image visually indistinguishable from a plaintext image. This novel substitution prevents cloud servers from detecting encrypted data, enabling the adoption of reversible data hiding (RDH) methods designed for plaintext images. The proposed framework offers several distinct advantages. Firstly, it ensures the confidentiality of sensitive information by concealing the linguistic content of the original image. Secondly, it supports reversible image editing, enabling the restoration of the original image from the encrypted version without any loss of data. Lastly, the integration of RDH techniques designed for plaintext images empowers the cloud server to embed supplementary data while preserving image quality. Incorporating convolutional neural network (CNN) and generative adversarial network (GAN) models, the framework ensures accurate data extraction and high-quality image restoration. The applications of this concealed knowledge are vast, spanning law enforcement, medical data privacy, and military communication. By addressing limitations of previous methods, it opens new avenues

关键词： Deep hiding Deep neural network GAN model Reversible data hiding (RDH)

来源：评论

学校读者我要写书评

暂无评论

Can LLMs Learn Uncertainty on Their Own? Expressing Uncertainty Effectively in A Self-Training Manner

Can LLMs Learn Uncertainty on Their Own? Expressing Uncertai...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Liu, Shudong Li, Zhaocong Liu, Xuebo Zhan, Runzhe Wong, Derek F. Chao, Lidia S. Zhang, Min NLP2CT Lab Department of Computer and Information Science University of Macau China Institute of Computing and Intelligence Harbin Institute of Technology Shenzhen China

ISBN: (纸本)9798891761643

Large language models (LLMs) often exhibit excessive, random, and uninformative uncertainty, rendering them unsuitable for decision-making in human-computer interactions. In this paper, we aim to instigate a heightened awareness of self-uncertainty in LLMs, enabling them to express uncertainty more effectively. To accomplish this, we propose an uncertainty-aware instruction tuning (UaIT) method, aligning LLMs' perception with the probabilistic uncertainty of the generation. We conducted experiments using LLaMA2 and Mistral on multiple free-form QA tasks. Experimental results revealed a surprising 45.2% improvement in the effectiveness of uncertainty expression by LLMs, accompanied by reasonably good out-of-domain generalization capabilities. Moreover, this uncertainty expression can serve as a valuable real-time basis for human decision-making, e.g., retrieving external documents and incorporating stronger LLMs. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Multiuser Computation Offloading for Long-Term Sequential Tasks in Mobile Edge computing Environments

引用

Tsinghua science and Technology 2023年第1期28卷 93-104页

作者： Huanhuan Xu Jingya Zhou Wenqi Wei Baolei Cheng School of Computer Science and Technology Soochow UniversitySuzhou 215006China State Key Laboratory of Mathematical Engineering and Advanced Computing Wuxi 214125China Provincial Key Laboratory for Computer Information Processing Technology Soochow UniversitySuzhou 215006China School of Computer Science Georgia Institute of TechnologyAtlantaGA 30332USA

Mobile edge computing has shown its potential in serving emerging latency-sensitive mobile applications in ultra-dense 5G networks via offloading computation workloads from the remote cloud data center to the nearby network ***,current computation offloading studies in the heterogeneous edge environment face multifaceted challenges:Dependencies among computational tasks,resource competition among multiple users,and diverse long-term *** applications typically consist of several functionalities,and one huge category of the applications can be viewed as a series of sequential *** this study,we first proposed a novel multiuser computation offloading framework for long-term sequential ***,we presented a comprehensive analysis of the task offloading process in the framework and formally defined the multiuser sequential task offloading ***,we decoupled the long-term offloading problem into multiple single time slot offloading problems and proposed a novel adaptive method to solve *** further showed the substantial performance advantage of our proposed method on the basis of extensive experiments.

关键词： mobile edge computing sequential tasks computation offloading dependency

来源：评论

学校读者我要写书评

暂无评论

Multi-Level Knowledge Engineering Approach for Mapping Implicit Aspects to Explicit Aspects

引用

computers, Materials & Continua 2022年第2期70卷 3491-3509页

作者： Jibran Mir Azhar Mahmood Shaheen Khatoon Computer Science Department Shaheed Zulfikar Ali Bhutto Institute of Science and Technology(SZABIST)Faculty of Computing and Engineering SciencesIslamabadPakistan Department of Computer Science Capital University of Science&TechnologyFaculty of ComputingIslamabad College of Computer Science and Information Technology King Faisal UniversitySaudi Arabia

Aspect’s extraction is a critical task in aspect-based sentiment analysis,including explicit and implicit aspects *** extensive research has identified explicit aspects,little effort has been put forward on implicit aspects extraction due to the complexity of the ***,existing research on implicit aspect identification is widely carried out on product reviews targeting specific aspects while neglecting sentences’dependency ***,in this paper,a multi-level knowledge engineering approach for identifying implicit movie aspects is *** proposed method first identifies explicit aspects using a variant of BiLSTM and CRF(Bidirectional Long Short Memory-Conditional Random Field),which serve as a memory to process dependent sentences to infer implicit *** can identify implicit aspects from four types of sentences,including independent and three types of dependent *** study is evaluated on a largemovie reviews dataset with 50k *** experimental results showed that the explicit aspect identification method achieved 89%F1-score and implicit aspect extraction methods achieved 76%*** addition,the proposed approach also performs better than the state-of-the-art techniques(NMFIAD andML-KB+)on the product review dataset,where it achieved 93%precision,92%recall,and 93%F1-score.

关键词： Movie NEs(named entities) aspects opinion words annotation process memory implicit aspects implicit aspects mapping word embedding and BiLSTM

来源：评论

学校读者我要写书评

暂无评论

SNR and RSSI Based an Optimized Machine Learning Based Indoor Localization Approach:Multistory Round Building Scenario over LoRa Network

引用

computers, Materials & Continua 2024年第8期80卷 1927-1945页

作者： Muhammad Ayoub Kamal Muhammad Mansoor Alam Aznida Abu Bakar Sajak Mazliham Mohd Su’ud Malaysian Institute of Information Technology(MIIT) Universiti Kuala LumpurKuala Lumpur50250Malaysia Faculty of Computing and Informatics Multimedia UniversityCyberjaya63100Malaysia Depertment of Computer Science DHA Suffa UniversityKarachiSindh75500Pakistan Riphah Institute of System Engineering(RISE) Faculty of ComputingRiphah International UniversityIslamabad46000Pakistan Malaysian France Institute(MFI) Universiti Kuala LumpurKuala Lumpur50250Malaysia Faculty of Engineering and Information Technology School of Computer ScienceUniversity of Technology SydneyUltimoNSW 2007Australia

In situations when the precise position of a machine is unknown,localization becomes *** research focuses on improving the position prediction accuracy over long-range(LoRa)network using an optimized machine learning-based *** order to increase the prediction accuracy of the reference point position on the data collected using the fingerprinting method over LoRa technology,this study proposed an optimized machine learning(ML)based *** signal strength indicator(RSSI)data from the sensors at different positions was first gathered via an experiment through the LoRa network in a multistory round layout *** noise factor is also taken into account,and the signal-to-noise ratio(SNR)value is recorded for every RSSI *** study concludes the examination of reference point accuracy with the modified KNN method(MKNN).MKNN was created to more precisely anticipate the position of the reference *** findings showed that MKNN outperformed other algorithms in terms of accuracy and complexity.

关键词： Indoor localization MKNN LoRa machine learning classification RSSI SNR localization

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing Brain Stroke Classification with Neuro-Xception and ELM Integration in Deep Learning 2

Revolutionizing Brain Stroke Classification with Neuro-Xcept...

引用

2nd International Conference on computing and Data Analytics, ICCDA 2024

作者： Jagadeesan, S. Ramalingam, Venkadeshan Abinash, P. Ilayaraja, V. Masood, Jafar Ali Ibrahim Syed Tamil Priya, D. School of Computer Science Engineering and Information Systems Vellore Institute of Technology Vellore India UTAS-Shinas branch Department of Computing and Information Sciences Oman School of Computer Science and Engineering Vellore Institute of Technology Vellore India

ISBN: (纸本)9798331542573

Stroke and related neurological disorders are difficult to classify accurately, hence new computational approaches are needed to enhance diagnosis. In order to transform the categorization of brain strokes, this study presents Neuro-Xception, a novel framework that combines the Xception architecture with Extreme Learning Machines (ELM). The rapid learning speed and enhanced generalisation capability of ELMs are combined with the feature extraction expertise of Xception to create Neuro-Xception. Through training on large neuroimaging datasets, Neuro-Xception learns complex patterns that are essential for differentiating between different types and intensities of stroke. By means of extensive testing on various datasets, Neuro-Xception exhibits notable gains in performance compared to traditional techniques, indicating its potential to revolutionise stroke detection in medical environments. Furthermore, Neuro-Xception's scalability and flexibility are enhanced by the combination of Xception and ELMs, which makes it a viable tool for real-time clinical decision support systems and personalised medicine. This study highlights how merging Xception and ELMs can revolutionise neurological healthcare and this development could lead to more accurate and efficient diagnostic techniques for treating stroke patients. Neuro-Xception offers a potential path for increasing diagnosis accuracy and expediting timely medical intervention. It is a major step forward in the use of artificial intelligence to enhance patient outcomes and advance the profession of stroke care. © 2024 IEEE.

关键词： Personalized medicine

来源：评论

学校读者我要写书评

暂无评论

SelectIT: Selective Instruction Tuning for LLMs via Uncertainty-Aware Self-Reflection 38

SelectIT: Selective Instruction Tuning for LLMs via Uncertai...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Liu, Liangxin Liu, Xuebo Wong, Derek F. Li, Dongfang Wang, Ziyi Hu, Baotian Zhang, Min Institute of Computing and Intelligence Harbin Institute of Technology Shenzhen China NLP2CT Lab Department of Computer and Information Science University of Macau China

Instruction tuning (IT) is crucial to tailoring large language models (LLMs) towards human-centric interactions. Recent advancements have shown that the careful selection of a small, high-quality subset of IT data can significantly enhance the performance of LLMs. Despite this, common approaches often rely on additional models or data, which increases costs and limits widespread adoption. In this work, we propose a novel approach, termed SelectIT, that capitalizes on the foundational capabilities of the LLM itself. Specifically, we exploit the intrinsic uncertainty present in LLMs to more effectively select high-quality IT data, without the need for extra resources. Furthermore, we introduce a curated IT dataset, the Selective Alpaca, created by applying SelectIT to the Alpaca-GPT4 dataset. Empirical results demonstrate that IT using Selective Alpaca leads to substantial model ability enhancement. The robustness of SelectIT has also been corroborated in various foundation models and domain-specific tasks. Our findings suggest that longer and more computationally intensive IT data may serve as superior sources of IT, offering valuable insights for future research in this area. Data, code, and scripts are freely available at https://***/Blue-Raincoat/SelectIT. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lay CO:Achieving Least Lossy Accuracy for Most Efficient RRAM-Based Deep Neural Network Accelerator via Layer-Centric Co-Optimization

引用

计算机科学技术学报（英文版） 2023年第2期38卷 328-347页

作者： Shao-Feng Zhao Fang Wang Bo Liu Dan Feng Yang Liu Wuhan National Laboratory for Optoelectronics Huazhong University of Science and TechnologyWuhan 430074China School of Computer Science and Technology Huazhong University of Science and TechnologyWuhan 430074China Cloud Computing and Big Data Institute Henan University of Economics and LawZhengzhou 450001China Wuhan National Laboratory for Optoelectronics Huazhong University of Science and TechnologyWuhan 430074China School of Computer Science and Technology Huazhong University of Science and TechnologyWuhan 430074China Research Institute of Huazhong University of Science and Technology in Shenzhen Shenzhen 518057China School of Computer and Artificial Intelligence Zhengzhou UniversityZhengzhou 450001China Wuhan National Laboratory for Optoelectronics Huazhong University of Science and TechnologyWuhan 430074China School of Computer Science and Technology Huazhong University of Science and TechnologyWuhan 430074China Cloud Computing and Big Data Institute Henan University of Economics and LawZhengzhou 450001China

Resistive random access memory(RRAM)enables the functionality of operating massively parallel dot prod-ucts and ***-based accelerator is such an effective approach to bridging the gap between Internet of Things devices'constrained resources and deep neural networks'tremendous *** to the huge overhead of Analog to Digital(A/D)and digital accumulations,analog RRAM buffer is introduced to extend the processing in analog and in *** analog RRAM buffer offers potential solutions to A/D conversion issues,the energy consumption is still challenging in resource-constrained environments,especially with enormous intermediate data ***,criti-cal concerns over endurance must also be resolved before the RRAM buffer could be frequently used in reality for DNN in-ference *** we propose LayCO,a layer-centric co-optimizing scheme to address the energy and endurance con-cerns altogether while strictly providing an inference accuracy *** relies on two key ideas:1)co-optimizing with reduced supply voltage and reduced bit-width of accelerator architectures to increase the DNN's error tolerance and achieve the accelerator's energy efficiency,and 2)efficiently mapping and swapping individual DNN data to a correspond-ing RRAM partition in a way that meets the endurance *** evaluation with representative DNN models demonstrates that LayCO outperforms the baseline RRAM buffer based accelerator by 27x improvement in energy effi-ciency(over TIMELY-like configuration),308x in lifetime prolongation and 6x in area reduction(over RAQ)while main-taining the DNN accuracy loss less than 1％.

关键词： deep learning accelerator resistive random access memory (RRAM) energy efficiency error tolerance

来源：评论

学校读者我要写书评

暂无评论

Optimized Deep Learning Framework for Personalized Nutritional Recommendations Across the Menstrual Cycle 5

Optimized Deep Learning Framework for Personalized Nutrition...

引用

5th International Conference on Sustainable Communication Networks and Application, ICSCNA 2024

作者： Logapriya, E. Surendran, R. Poornima, D. Kartheesan, L. Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Chennai India Sathyabama Institute of Science and Technology School of Computing Department of Computer Science and Engineering Chennai India Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Department of Computer Science and Engineering Avadi Chennai India

ISBN: (数字)9798331530013

ISBN: (纸本)9798331530013

This study presents a deep learning architecture designed to monitor a woman's menstrual cycle and provide personalized dietary suggestions to improve health. The system employs complex models, including CNNs, RNNs, Transformers, T-GNNs, VAEs, and GANs, to predict the intensity of symptoms, determine the phases of the menstrual cycle, and offer personalized nutritional advice. By analysing data from menstrual diaries, food consumption, and symptom tracking, the method establishes a strong association between hormonal fluctuations, symptoms, and nutritional needs. Model accuracy is enhanced through data preprocessing techniques, including imputation, normalization, and augmentation, which provide high-quality training inputs. Research shows that Transformer models outperform other approaches in phase recognition (96% accuracy) and phase-aware nutritional recommendation accuracy. This research highlights how deep learning can transform women's healthcare by providing personalized nutritional recommendations based on data that align with the phases of a woman's menstrual cycle, ultimately contributing to improved health for women. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Custom Transformer-Based Approach for Enhanced Bengali Automatic Speech Recognition 1

Custom Transformer-Based Approach for Enhanced Bengali Autom...

引用

1st International Conference on Innovations in Communications, Electrical and computer Engineering, ICICEC 2024

作者： Deepa, R. Jayalakshmi, V. Thilakavathy, P. Manikandan, G. Surendran, R. SRM Institute of Science and Technology School of Computing Department of Computing Technologies Chennai India SRM Institute of Science and Technology Department of Computer Science and Applications Ramapuram Chennai India Vels Institute of Science and Engineering Vels Institute of Science Technology and Advanced Studies Chennai India Anna University St. Joseph's College of Engineering Department of Artificial Intelligence and Machine Learning Chennai India Saveetha Institute of Medical and Technical Sciences Saveetha School of Engineering Department of Computer Science and Engineering Chennai India

ISBN: (数字)9798350376517

ISBN: (纸本)9798350376517

Automated speech recognition (ASR) systems struggle with Bengali, which is the fifth most spoken language. Bengali has a varied morphology, many dialects, and limited high-quality annotated voice data. Traditional voice recognition techniques work well however struggle with Bengali phonetics and syntax. This method eliminates the need for large amounts of tagged data and the difficulties of differentiating sounds in different languages, both issues are addressed by this procedure. The research in this paper offers an Intellectual Bengali Speech Recognition System using Deep Learning Techniques (IBSRS-DLT). The model in this system uses transformer architecture and attention processes to manage Bengali speech pattern long-range dependencies. The IBSRS-DLT outperforms standard models in recognition accuracy using a pre-trained transformer model modified for phonetic and lexical correctness and fine-tuned on Bengali datasets. After training on a vast Bengali speech corpus, real-world datasets and rigorous simulations confirm the system. IBSRS-DLT can be used for automatic transcription, voice-controlled devices, and Bengali virtual assistants, such applications are efficient and accessible. The simulation showed a considerable boost in recognition speed and a decrease in keyword mistakes compared to current methods. This paper shows that deep learning can solve linguistic issues in ASR systems, enabling scalable regional language detection. © 2024 IEEE.

关键词： Deep learning

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：