检索结果-内蒙古大学图书馆

IEEE International Conference on Communications (ICC)

作者： Cheuk Ting Leung Rohan Ghosh Mehul Motani Department of Electrical & Computer Engineering National University of Singapore Institute of Data Science Institute for Digital Medicine N.1 Institute for Health

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

Conventional communication systems focus on re-covering the messages sent by the transmitter at the receiver, by undoing the errors introduced by the channel. In semantic communication, the goal is instead to preserve the semantics (i.e., meaning) of the intended message $X$ . A well-known framework adopted by the literature in modelling semantic communication is task-oriented communication. There, an encoder generates a feature representation $Z$ (e.g., a trained neural network) to transmit across the channel to fulfill a certain task (e.g., classification with labels $Y$ ). This naturally leads to the study of the information bottleneck (IB) principle - which proposes reducing the mutual information $I (X; Z)$ (compression) while simultaneously maxi-mizing $I (Z; Y)$ (fitting). Although this objective seems initially meaningful, we posit that both the standard task-oriented setting for studying semantic communication and the IB approach may not be optimal. In this paper, we first propose a novel paradigm of multi-task generalizable communication, where, at test time, the messages may originate from a different unseen classification task compared to those during training. We intuitively reason that the multi-task generalizable setting is more feasible and relevant in the context of real-world semantic communication, compared to task-oriented communication. Next, we propose a Reconstruction Loss Aware (RLA) approach, which yields better feature encodings that are more generalizable towards unseen tasks. We empirically observe that, in the proposed paradigm, the IB principle is not optimal. We also demonstrate that RLA is better (15%-20% higher accuracy) compared to a benchmark approach for task-oriented communication, for diverse non-overlapping unseen tasks.

关键词： Training Transmitters Semantics Neural networks Receivers Benchmark testing Multitasking

来源：评论

学校读者我要写书评

暂无评论

Parallelization of Software Systems Test Case Selection Algorithm Based on Singular Value Decomposition

arXiv

引用

arXiv 2022年

作者： Moghaddam, Mahdi Movahedian Department of Data and Computer Science Faculty of Mathematical Science Shahid Beheshti University Tehran Iran

When developing a software system, a change in one part of the system may lead to unwanted changes in other parts of the system. These affected parts may interfere with system performance, so regression testing is used to deal with these disorders. This test seeks to re-measure these sections to prevent these abnormalities, but it is difficult to identify these sections for re-examination. We try to cluster the changes of our software system based on the system functions by singular value decomposition, to be able to use to identify these parts during a new change, to perform the test again. In order to increase speedup, our calculations were performed in parallel on shared memory systems so that by increasing the scale of software systems, an optimal answer could be obtained. © 2022, CC BY.

关键词： Singular value decomposition

来源：评论

学校读者我要写书评

暂无评论

Enabling Efficient NVM-Based Text Analytics without Decompression

Enabling Efficient NVM-Based Text Analytics without Decompre...

引用

International Conference on data Engineering

作者： Xiaokun Fang Feng Zhang Junxiang Nong Mingxing Zhang Puyun Hu Yunpeng Chai Xiaoyong Du Key Laboratory of Data Engineering and Knowledge Engineering (MOE) and School of Information Renmin University of China Department of Computer Science and Engineering Tsinghua University

ISBN: (数字)9798350317152

ISBN: (纸本)9798350317169

Text analytics directly on compression (TADOC) is a promising technology designed for handling big data analytics. However, a substantial amount of DRAM is required for high performance, which limits its usage in many important scenarios where the capacity of DRAM is limited, such as memory-constrained systems. Non-volatile memory (NVM) is a novel storage technology that combines the advantage of reading per-formance and byte addressability of DRAM with the durability of traditional storage devices like SSD and HDD. Unfortunately, no research demonstrates how to use NVM to reduce DRAM utilization in compressed data analytics. In this paper, we propose N-TADOC, which substitutes DRAM with NVM while maintaining TADOC's analytics performance and space savings. Utilizing an NVM block device to reduce DRAM utilization presents two challenges, including poor data locality in traversing datasets and auxiliary data structure reconstruction on NVM. We develop novel designs to solve these challenges, including a pruning method with NVM pool management, bottom-up upper bound estimation, correspondent data structures, and persistence strategy at different levels of cost. Experimental results show that on four real-world datasets, N-TADOC achieves 2.04× performance speedup compared to the processing directly on the uncompressed data and 70.7% DRAM space saving compared to the original TADOC.

关键词： Performance evaluation Upper bound data analysis Costs Nonvolatile memory Random access memory Estimation

来源：评论

学校读者我要写书评

暂无评论

Layer Attack Unlearning: Fast and Accurate Machine Unlearning via Layer Level Attack and Knowledge Distillation

arXiv

引用

arXiv 2023年

作者： Kim, Hyunjune Lee, Sangyong Woo, Simon S. Department of Applied Data Science Sungkyunkwan University Korea Republic of Department of Artificial Intelligence Sungkyunkwan University Korea Republic of Department of Computer Science and Engineering Sungkyunkwan University Korea Republic of

Recently, serious concerns have been raised about the privacy issues related to training datasets in machine learning algorithms when including personal data. Various regulations in different countries, including the GDPR, grant individuals to have personal data erased, known as ‘the right to be forgotten’ or ‘the right to erasure’. However, there has been less research on effectively and practically deleting the requested personal data from the training set while not jeopardizing the overall machine learning performance. In this work, we propose a fast and novel machine unlearning paradigm at the layer level called layer attack unlearning, which is highly accurate and fast compared to existing machine unlearning algorithms. We introduce the Partial-PGD algorithm to locate the samples to forget efficiently. In addition, we only use the last layer of the model inspired by the Forward-Forward algorithm for unlearning process. Lastly, we use Knowledge Distillation (KD) to reliably learn the decision boundaries from the teacher using soft label information to improve accuracy performance. We conducted extensive experiments with SOTA machine unlearning models and demonstrated the effectiveness of our approach for accuracy and end-to-end unlearning performance. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

Prediction of Kinase-Substrate Associations Using The Functional Landscape of Kinases and Phosphorylation Sites 28

Prediction of Kinase-Substrate Associations Using The Functi...

引用

28th Pacific Symposium on Biocomputing, PSB 2023

作者： Ayati, Marzieh Yilmaz, Serhan Blasco Tavares Pereira Lopes, Filipa Chance, Mark Koyuturk, Mehmet Department Of Computer Science University Of Texas Rio Grande Valley EdinburgTX United States Department Of Computer And Data Sciences ClevelandOH United States Department Of Nutrition ClevelandOH United States Center For Proteomics And Bioinformatics ClevelandOH United States Case Comprehensive Cancer Center Case Western Reserve University ClevelandOH United States

ISBN: (纸本)9789811270604

Protein phosphorylation is a key post-translational modification that plays a central role in many cellular processes. With recent advances in biotechnology, thousands of phosphorylated sites can be identified and quantified in a given sample, enabling proteome-wide screening of cellular signaling. However, for most (> 90%) of the phosphorylation sites that are identified in these experiments, the kinase(s) that target these sites are unknown. To broadly utilize available structural, functional, evolutionary, and contextual information in predicting kinase-substrate associations (KSAs), we develop a network-based machine learning framework. Our framework integrates a multitude of data sources to characterize the landscape of functional relationships and associations among phosphosites and kinases. To construct a phosphosite-phosphosite association network, we use sequence similarity, shared biological pathways, co-evolution, co-occurrence, and co-phosphorylation of phosphosites across different biological states. To construct a kinase-kinase association network, we integrate protein-protein interactions, shared biological pathways, and membership in common kinase families. We use node embeddings computed from these heterogeneous networks to train machine learning models for predicting kinase-substrate associations. Our systematic computational experiments using the PhosphositePLUS database shows that the resulting algorithm, NetKSA, outperforms two state-of-the-art algorithms, including KinomeXplorer and LinkPhinder, in overall KSA prediction. By stratifying the ranking of kinases, NetKSA also enables annotation of phosphosites that are targeted by relatively less-studied kinases. Availability: The code and data are available at ***/NetKSA/. © 2022 The Authors.

关键词： Association reactions

来源：评论

学校读者我要写书评

暂无评论

High Speed and Low Cost One-to-Many VLC Using Polymer-Dispersed Liquid Crystals

High Speed and Low Cost One-to-Many VLC Using Polymer-Disper...

引用

IEEE International Conference on Communications (ICC)

作者： Yuehu Jiang Haonan Wu Yi-Chao Chen Guangtao Xue Department of Computer Science and Technology Shanghai JiaoTong University China Shanghai Key Laboratory of Trusted Data Circulation and Governance in Web3

ISBN: (数字)9781728190549

ISBN: (纸本)9781728190556

Visible light communication (VLC) is considered a solution to the scarcity of radio frequency communication resources due to its abundant spectrum resources and rapid intensity modulation capability. It has a wide range of applications in indoor positioning and intelligent transport systems. For example, in Connected and Autonomous Vehicle scenarios, VLC uses traffic lights to warn vehicles at different distances in abnormal situations, thus preventing potential traffic accidents. To facilitate fast, long-range VLC communication in such one-to-many communication scenarios, current systems typically use optical cameras or digital micro-mirror devices as receivers. However, there are several challenges associated with these devices. Optical cameras have a limited sampling rate, resulting in reduced effective throughput. Other receivers, such as digital micro-mirror devices, are relatively costly, which hinders their widespread use. In this paper, we propose a novel, low-cost, and high-speed VLC scheme. We use a low-cost material called Polymer-Dispersed Liquid Crystal as the measurement matrix, reducing the cost by 99% compared to digital micro-mirror devices. We implement hierarchical coding based on compressive sensing to reduce data redundancy and thus improve communication throughput. Empirical experiments conducted using four pho-to diodes at the receiver show a 120% improvement in overall throughput compared to existing one-to-many VLC systems.

关键词： Integrated optics Costs Throughput Optical receivers Encoding Optical sensors Liquid crystals

来源：评论

学校读者我要写书评

暂无评论

Persistent Classification: A New Approach to Stability of data and Adversarial Examples

arXiv

引用

arXiv 2024年

作者： Bell, Brian Geyer, Michael Glickenstein, David Hamm, Keaton Scheidegger, Carlos Fernandez, Amanda Moore, Juston Department of Mathematics University of Arizona Los Alamos National Laboratory United States Department of Computer Science University of Texas San Antonio Los Alamos National Laboratory United States Department of Mathematics University of Arizona United States Division of Data Science University of Texas Arlington United States Department of Computer Science University of Arizona United States Department of Computer Science University of Texas San Antonio United States Los Alamos National Laboratory United States

There are a number of hypotheses underlying the existence of adversarial examples for classification problems. These include the high-dimensionality of the data, high codimension in the ambient space of the data manifolds of interest, and that the structure of machine learning models may encourage classifiers to develop decision boundaries close to data points. This article proposes a new framework for studying adversarial examples that does not depend directly on the distance to the decision boundary. Similarly to the smoothed classifier literature, we define a (natural or adversarial) data point to be (γ, σ)-stable if the probability of the same classification is at least γ for points sampled in a Gaussian neighborhood of the point with a given standard deviation σ. We focus on studying the differences between persistence metrics along interpolants of natural and adversarial points. We show that adversarial examples have significantly lower persistence than natural examples for large neural networks in the context of the MNIST and ImageNet datasets. We connect this lack of persistence with decision boundary geometry by measuring angles of interpolants with respect to decision boundaries. Finally, we connect this approach with robustness by developing a manifold alignment gradient metric and demonstrating the increase in robustness that can be achieved when training with the addition of this metric. © 2024, CC BY.

关键词： Interpolation

来源：评论

学校读者我要写书评

暂无评论

Dynamic Feature Selection Algorithm for Enhanced Time-Series Energy Consumption Forecasting

Dynamic Feature Selection Algorithm for Enhanced Time-Series...

引用

Integration of Emerging Technologies for the Digital World (ICIETDW), International Conference on

作者： K. Saranya Veeramalai Sankaradass V. Antony Suresh G. Manimozhi Department of Artificial Intelligence and Data Science Panimalar Institute of Technology Chennai India Department of Computer Science and Engineering Chennai Institute of Technology Chennai India Department of Information Technology St. Joseph College of Engineering Chennai India

ISBN: (数字)9798331541170

ISBN: (纸本)9798331541187

Accurate power intake projecting is essential for enhancing power source administration, lowering power expenses and also preparing for future power demands. Because of solar power's intermittent nature coupled with minimal ability to handle solar problems grid drivers worldwide have a hard time to efficiently incorporate solar power right into the system to meet present need. This requires a thorough service to properly anticipate as well as take care of solar energy radiation as well as generation, attending to the progressing characteristics of supply as well as need on the grid. The purpose of the job is to produce a Artificial Intelligence (AI) design for exact forecast of solar energy radiation coupled with generation, along with improved power intake projecting by utilizing vibrant attributes to enhance flexibility and also precision. The method includes numerous artificial intelligence formulas from monitored set understanding coupled with deep discovering groups for maximized projecting. The datasets for radiation as well as generation have actually been sourced from Kaggle's system. The very best precision has actually been acquired from the Linear Regression for radiation is 97.75 together with the Random Forest Regression for generation is *** employing SARIMAX, the system aims to capture the complex temporal dynamics inherent in energy consumption data, thereby enabling precise forecasting of future consumption patterns. The optimized forecasting model facilitates improved decision-making for solar power generation scheduling, resource allocation, and grid integration, ultimately contributing to the advancement of sustainable energy systems.

关键词： Energy consumption Accuracy Heuristic algorithms Solar energy Predictive models Feature extraction Dynamic scheduling Solar power generation Resource management Forecasting

来源：评论

学校读者我要写书评

暂无评论

Internet of Things (IoT): Trends, Challenges, Simulators, Emulators and Test-beds

Internet of Things (IoT): Trends, Challenges, Simulators, Em...

引用

International Conference on Computing Communication Control and Automation (ICCUBEA)

作者： Jameel Shehu Yalli Mohd Hilmi Hasan Alawi Al-Qushaibi Dahiru Adamu Aliyu Saipunidzam Mahamad Computer & Information Sciences Department Universiti Teknologi PETRONAS Perak Malaysia Centre for Research in Data Science Universiti Teknologi PETRONAS Perak Malaysia

ISBN: (数字)9798350391770

ISBN: (纸本)9798350391787

During the past few years, the deployment and adoption of IoT in the world have increased significantly. The growth rate of IoT increases annually. Researchers today rate the security of IoT to determine its acceptance trend by the public. This paper answers five research questions that show the theme trends for the IoT development, the main challenges of IoT security, and the choice of simulators, emulators, and test-beds used to implement IoT scenarios. A comparison of the available tools/software/simulators that could run the IoT model is presented to give users insight to visualize the working concept of the IoT.

关键词： Visualization Automation Computational modeling Market research Security Internet of Things

来源：评论

学校读者我要写书评

暂无评论

Deepfake Video Detection using Neural Networks 1

Deepfake Video Detection using Neural Networks

引用

1st IEEE International Conference on Blockchain and Distributed Systems Security, ICBDS 2022

作者： Jiwtode, Mayur Asati, Aman Kamble, Shailesh Damahe, Lalit Yeshwantrao Chavan College of Engineering Department of Computer Technology Nagpur India Indira Gandhi Delhi Technical University for Women Department of Artificial Intelligence and Data Science Delhi India Yeshwantrao Chavan College of Engineering Head of Department of Computer Technology Nagpur India

ISBN: (数字)9781665428323

ISBN: (纸本)9781665428323

Over the past few years, free mobile application tools on Artificial Intelligence and deep learning have made it easy to create reliable face exchanges in a video called 'DeepFake' (DF) video that leaves a little hint of traces to check if its fake. Creating a computerized edited video has been demonstrated for quite a long time by actually taking advantage of enhanced visual effects. Recently, Artificial Intelligence has led to a rise in fake content and the ability to access free tools to create it. These purported AI-engineered media are normally called DeepFake(DF). Making a DF with a computerized AI tool is a simple job. However, with regards to identifying this DF, it's a major challenging dispute. Preparing the calculations and training the model to distinguish DF is difficult. The challenge to train an algorithm model to spot the DeepFake (DF) is not simple. We have tried recognizing DF with the use of CNN and RNN. The framework utilizes a CNN for feature extraction at the frame level. The model uses features extracted from the frame level to train the RNN, which then learns to classify videos according to their temporal inconsistencies. Anticipated results, when compared with a large number of hoax videos, were gathered from standard datasets. Using a simple architecture, we will show you how this errand can make your framework accurate. © 2022 IEEE.

关键词： Large dataset

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：