检索结果-内蒙古大学图书馆

Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA)

作者： Sho Inoue Kun Zhou Shuai Wang Haizhou Li School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen China Speech Lab Institute for Intelligent Computing Alibaba Group Singapore

ISBN: (数字)9798350367331

ISBN: (纸本)9798350367348

It remains a significant challenge how to quantitatively control the expressiveness of speech emotion in speech generation. In this work, we propose an approach for quantitative manipulation of the emotion rendering for emotion editing in speech generation. We apply a hierarchical emotion distribution extractor, i.e. Hierarchical ED, that quantifies the intensity of emotions at different levels of granularity. Hierarchical ED is subsequently integrated into the FastSpeech2 framework, guiding the model to learn emotion intensity at phoneme, word, and utterance levels. During synthesis, users can manually edit the emotional intensity of the generated voices. Both objective and subjective evaluations demonstrate the effectiveness of the proposed network in terms of fine-grained quantitative emotion editing.

关键词： Asia Information processing Rendering (computer graphics) Text to speech Speech processing

来源：评论

学校读者我要写书评

暂无评论

Disease Detection in Arecanut using Convolutional Neural Network

Disease Detection in Arecanut using Convolutional Neural Net...

引用

Advances in computing, Communication and Applied Informatics (ACCAI), International Conference on

作者： V Arun Karthik Jatin Shivaprakash D Rajeswari Department of Data Science and Business Systems School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai India

ISBN: (数字)9798350389449

ISBN: (纸本)9798350389456

Arecanut, also known as betel nut, is a tropical crop predominantly grown in India. The country holds the second position globally in terms of the production and consumption of arecanut. Throughout its life cycle, the areca nut plant is susceptible to a variety of diseases affecting its roots, trunk, leaves, and fruits. Some of these diseases are visible to the naked eye, while others are not. Farmers traditionally analyze every crop to detect any signs of disease, a process that is extremely *** this study, we propose an automatic system that aids in detecting the diseases of arecanut leaves using Convolutional Neural Networks (CNN) and deep learning algorithms. A Convolutional Neural Network is a Deep Learning algorithm that takes an image as input, assigns learnable weights and biases to various features in the image, and then learns from the results to distinguish one feature from another. This approach significantly reduces the manual effort required in disease detection and increases the efficiency of the process. A datset is developed with many photos of healthy and sick arecanut leaves in order to train and evaluate the CNN model. An 80:20 split of the dataset was made into training and testing data. The goal of training the model over multiple epochs was to maximize validation and test accuracy while minimizing loss. It was discovered that the suggested method for diagnosing arecanut illnesses was precise and successful. This study contributes to the ongoing efforts to leverage technology in agriculture, with the aim of improving crop health and productivity. This study indicate that the proposed system can accurately identify diseases in arecanut leaves with a good and best accuracy. The system was able to distinguish between healthy and diseased leaves, and suggest appropriate remedies for the detected diseases. These findings demonstrate the potential of using Convolutional Neural Networks and deep learning algorithms in the field of agriculture, par

关键词： Deep learning Training Productivity Plant diseases Accuracy Crops Agriculture

来源：评论

学校读者我要写书评

暂无评论

Deepfake Detection System

Deepfake Detection System

引用

Networking and Communications (ICNWC), International Conference on

作者： Shaswat Shrivas Aditya Rai M. Lakshmi Department of Data Science and Business Systems School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai India

ISBN: (数字)9798350365269

ISBN: (纸本)9798350365276

Detecting deepfake content presents a formidable challenge, necessitating advanced methodologies. This paper proposes a holistic strategy employing Facenet_pytorch, MTCNN, and InceptionResnetV1 for robust deepfake detection. Facenet_pytorch serves as the cornerstone for facial feature recognition, while MTCNN efficiently identifies faces in images during preprocessing. InceptionResnetV1 scrutinizes visual details to detect subtle anomalies indicative of manipulation, such as unnatural facial expressions and incongruent lip synchronization. Our approach underscores the importance of maintaining a delicate balance between accurate detection and individual privacy rights. By leveraging these advanced models, we achieve significant strides in differentiating manipulated content from authentic media, contributing to the ethical deployment of deepfake detection technologies.

关键词： Deepfakes Visualization Social networking (online) Law enforcement Forensics Entertainment industry Biodiversity

来源：评论

学校读者我要写书评

暂无评论

Experimental Analysis on IVPM Implementation for Image Processing Workflows

Experimental Analysis on IVPM Implementation for Image Proce...

引用

Advances in computing, Communication and Applied Informatics (ACCAI), International Conference on

作者： Sidhant Sharma Md Anas Hasan Abhivansh Bhatnagar Valadri Pardhavan Reddy K. Sornalakshmi Department of Data Science and Business Systems School of Computing Faculty of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai India

ISBN: (数字)9798350389449

ISBN: (纸本)9798350389456

IPVM (Interplanetary Virtual Machine) is a versatile protocol, an extension of IPFS capable of handling a diverse range of computations, spanning from small-scale to large-scale tasks. It plays a pivotal role in driving the transition towards open decentralized computing infrastructure, shifting from conventional centralized cloud services. By making use of IPFS content addressing, IPVM streamlines computation processes, facilitating distributed scheduling, caching, memoization, and adaptive optimization. This exploration highlights the emergence of a decentralized ecosystem promising enhanced resilience, scalability, and inclusivity in computing frameworks.

关键词： Processor scheduling Scalability Image processing Ecosystems Streaming media InterPlanetary File System Task analysis

来源：评论

学校读者我要写书评

暂无评论

Speech Emotion Recognition Using data Augmentation

Speech Emotion Recognition Using Data Augmentation

引用

Advances in computing, Communication and Applied Informatics (ACCAI), International Conference on

作者： Tanisha Kapoor Arnaja Ganguly D Rajeswari Department of Data Science and Business Systems School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chennai India

ISBN: (数字)9798350389449

ISBN: (纸本)9798350389456

Speech emotion recognition (SER) is a vital aspect of human-computer interaction systems, enabling applications such as virtual assistants, affective computing, and mental health monitoring. In this work, deep learning methods to audio data to suggest a fresh approach to SER are applied. While still in its infancy, CNN is capable of accurately identifying emotions in human speech, but it requires careful study in order to produce reliable *** studywas done using a heterogeneous dataset that included recordings of different emotional states from many sources, such as TESS, RAVDESS, SAVEE, CREMAD, and others. A deep neural network model was built with many fully connected layers and ReLU activation functions after preprocessing and feature extraction. The suggested method’ s effectiveness was assessed using common performance indicators like F1-score, recall, accuracy, and precision. Our tests produced encouraging outcomes, with an accuracy of 84% on the test dataset. This demonstrates how well the suggested technique works to discern emotions from speech audio information.

关键词： Emotion recognition Accuracy Virtual assistants Speech recognition Mental health Feature extraction Recording

来源：评论

学校读者我要写书评

暂无评论

Achilles’ Heels: Vulnerable Record Identification in Synthetic data Publishing

arXiv

引用

arXiv 2023年

作者： Meeus, Matthieu Guépin, Florent Creţu, Ana-Maria de Montjoye, Yves-Alexandre Department of Computing and Data Science Institute Imperial College London London United Kingdom

Synthetic data is seen as the most promising solution to share individual-level data while preserving privacy. Shadow modeling-based Membership Inference Attacks (MIAs) have become the standard approach to evaluate the privacy risk of synthetic data. While very effective, they require a large number of datasets to be created and models trained to evaluate the risk posed by a single record. The privacy risk of a dataset is thus currently evaluated by running MIAs on a handful of records selected using ad-hoc methods. We here propose what is, to the best of our knowledge, the first principled vulnerable record identification technique for synthetic data publishing, leveraging the distance to a record’s closest neighbors. We show our method to strongly outperform previous ad-hoc methods across datasets and generators. We also show evidence of our method to be robust to the choice of MIA and to specific choice of parameters. Finally, we show it to accurately identify vulnerable records when synthetic data generators are made differentially private. The choice of vulnerable records is as important as more accurate MIAs when evaluating the privacy of synthetic data releases, including from a legal perspective. We here propose a simple yet highly effective method to do so. We hope our method will enable practitioners to better estimate the risk posed by synthetic data publishing and researchers to fairly compare ever improving MIAs on synthetic data. Copyright © 2023, The Authors. All rights reserved.

关键词： Risk perception

来源：评论

学校读者我要写书评

暂无评论

Synthetic is all you need: removing the auxiliary data assumption for membership inference attacks against synthetic data

arXiv

引用

arXiv 2023年

作者： Guépin, Florent Meeus, Matthieu Creţu, Ana-Maria de Montjoye, Yves-Alexandre Department of Computing and Data Science Institute Imperial College London London United Kingdom

Synthetic data is emerging as one of the most promising solutions to share individual-level data while safeguarding privacy. While membership inference attacks (MIAs), based on shadow modeling, have become the standard to evaluate the privacy of synthetic data, they currently assume the attacker to have access to an auxiliary dataset sampled from a similar distribution as the training dataset. This is often seen as a very strong assumption in practice, especially as the proposed main use cases for synthetic tabular data (e.g. medical data, financial transactions) are very specific and don’t have any reference datasets directly available. We here show how this assumption can be removed, allowing for MIAs to be performed using only the synthetic data. Specifically, we developed three different scenarios: (S1) Black-box access to the generator, (S2) only access to the released synthetic dataset and (S3) a theoretical setup as upper bound for the attack performance using only synthetic data. Our results show that MIAs are still successful, across two real-world datasets and two synthetic data generators. These results show how the strong hypothesis made when auditing synthetic data releases – access to an auxiliary dataset – can be relaxed, making the attacks more realistic in practice. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Logical Relation Modeling and Mining in Hyperbolic Space for Recommendation 40

Logical Relation Modeling and Mining in Hyperbolic Space for...

引用

40th IEEE International Conference on data Engineering, ICDE 2024

作者： Tan, Yanchao Lv, Hang Zhou, Zihao Guo, Wenzhong Xiong, Bo Liu, Weiming Chen, Chaochao Wang, Shiping Yang, Carl College of Computer and Data Science Fuzhou University Fuzhou China Engineering Research Center of Big Data Intelligence Ministry of Education Fuzhou China Fujian Key Laboratory of Network Computing and Intelligent Information Processing Fuzhou University Fuzhou China Institute for Artificial Intelligence University of Stuttgart Stuttgart Germany College of Computer Science Zhejiang University Hangzhou China Emory University Department of Computer Science Atlanta United States

ISBN: (纸本)9798350317152

The sparse interactions between users and items have aggravated the difficulty of their representations in recommender systems. Existing methods leverage tags to alleviate the sparsity problem but ignore prevalent logical relations among items and tags (e.g., membership, hierarchy, and exclusion), which can be leveraged to enhance the accuracy of modeling user preferences and conducting recommendations. To this end, we propose to extract logical relations among item tags from existing tag taxonomies and exploit the individual strengths of the Poincaré and the Lorentz models in hyperbolic space for logical relation modeling towards enhanced recommendations. Moreover, we find that the logical relations directly extracted from existing tag taxonomies can be inaccurate and coarse. Therefore, we further devise innovative consistency-based and granularity-based weighting mechanisms based on user behavior patterns for data-driven logical relation mining that can be jointly optimized along with recommendations in an end-to-end fashion. Extensive experiments on four real-world benchmark datasets show drastic performance gains brought by our proposed framework, which constantly achieves an average of 8.25% improvement over state-of-the-art competitors regarding both Recall and NDCG metrics. Insightful case studies further demonstrate that our automatically refined logical relations are highly accurate and interpretable. © 2024 IEEE.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Improved Memory Type Product Estimator for Population Mean in Stratified Random Sampling Under Linear Cost Function

引用

SN Computer science 2023年第3期4卷 1-9页

作者： Yadav, Subhash Kumar Vishwakarma, Gajendra K. Varshney, Rahul Pal, Arun Department of Statistics School of Physical and Decision Sciences Babasaheb Bhimrao Ambedkar University Lucknow 226025 India Department of Mathematics and Computing Indian Institute of Technology Dhanbad Dhanbad 826004 India Institute of Actuarial Science and Data Analytics UCSI University Kuala Lumpur 56000 Malaysia

In this article, we consider the memory type product estimator to estimate the population mean of the study variable in stratified random sampling. The suggested estimators’ bias and mean square error (MSE) expressions for the first order of approximation are derived. The proposed estimator is compared with the competing estimators under consideration and the efficiencies’ conditions are obtained. Further, we obtain the optimum allocation of samples in different strata along with MSEs. To achieve the solution of the problem, we frame our problem as AINLPP and solve through genetic programming technique. To show the efficiency of the estimator, we conduct simulation study. An empirical analysis based on real-life data is also offered to support the findings of the simulation study. The most efficient estimator is recommended for different practical applications. The suggested estimator is proven to be the best among the competing estimators. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： AINLPP Genetic programming technique Linear cost function Product estimator Stratified random sampling

来源：评论

学校读者我要写书评

暂无评论

An Improved Finite-time Analysis of Temporal Difference Learning with Deep Neural Networks 41

An Improved Finite-time Analysis of Temporal Difference Lear...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Ke, Zhifa Wen, Zaiwen Zhang, Junyu Center for Data Science Peking University China Beijing International Center for Mathematical Research Center for Machine Learning Research Changsha Institute for Computing and Digital Economy Beijing China Department of Industrial Systems Engineering and Management National University of Singapore Singapore

Temporal difference (TD) learning algorithms with neural network function parameterization have well-established empirical success in many practical large-scale reinforcement learning tasks. However, theoretical understanding of these algorithms remains challenging due to the nonlinearity of the action-value approximation. In this paper, we develop an improved non-asymptotic analysis of the neural TD method with a general L-layer neural network. New proof techniques are developed and an improved new Õ(ϵ-1) sample complexity is derived. To our best knowledge, this is the first finite-time analysis of neural TD that achieves an Õ(ϵ-1) complexity under the Markovian sampling, as opposed to the best known Õ(ϵ-2) complexity in the existing literature. Copyright 2024 by the author(s)

关键词： Adversarial machine 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：