检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Li, Hongru Yu, Wentao He, Hengtao Shao, Jiawei Song, Shenghui Zhang, Jun Letaief, Khaled B. Dept. of Electronic and Computer Engineering The Hong Kong University of Science and Technology Hong Kong

Task-oriented communication is an emerging paradigm for next-generation communication networks, which extracts and transmits task-relevant information, instead of raw data, for downstream applications. Most existing deep learning (DL)-based task-oriented communication systems adopt a closed-world scenario, assuming either the same data distribution for training and testing, or the system could have access to a large out-of-distribution (OoD) dataset for retraining. However, in practical open-world scenarios, task-oriented communication systems need to handle unknown OoD data. Under such circumstances, the powerful approximation ability of learning methods may force the task-oriented communication systems to overfit the training data (i.e., in-distribution data) and provide overconfident judgments when encountering OoD data. Based on the information bottleneck (IB) framework, we propose a class conditional IB (CCIB) approach to address this problem in this paper, supported by information-theoretical insights. The idea is to extract distinguishable features from in-distribution data while keeping their compactness and informativeness. This is achieved by imposing the class conditional latent prior distribution and enforcing the latent of different classes to be far away from each other. Simulation results shall demonstrate that the proposed approach detects OoD data more efficiently than the baselines and state-of-the-art approaches, without compromising the rate-distortion tradeoff. Copyright © 2023, The Authors. All rights reserved.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

A Novel Hybrid Deep Learning Model for Detecting Breast Cancer

TechRxiv

引用

TechRxiv 2023年

作者： Banerjee, Avishek Chakraborty, Debasis Bhattacharyya, Basanti Nath, Ira Dey, Debashis Garg, Deepak Dept. of Computer Science and Business Systems Asansol Engineering College West Bengal 713305 India Dept. of Computer Science and Engineering Asansol Engineering College Asansol713305 India Dept. of Computer Science and Engineering Pailan College of Management and Technology West Bengal 700139 India Dept. of Computer Science and Engineering JIS College of Engineering West Bengal 741235 India Dept. of Computer Science and Engineering Maulana Abul Kalam Azad University of Technology West Bengal 700064 India School of Computer Science and Artificial Intelligence SR University Telangana Warangal506371 India

Cancer is a deformity of the body cells that grow out of control and spread to other parts of body. According to the American Cancer Society, early identification of cancer resulted in a 99% chance of survival in the localized stage. In this article, a benchmark cancer dataset is considered for the purpose of investigation. After data preprocessing, we attempt to predict cancer type using our proposed deep learning model named as CNN-Soft-T that combines the feature selection (FS) and classification. FS is introduced using histogram technique for anomaly detection and dimension reduction. We have compared the performance of this model to other cutting-edge machine learning approaches including Logistic Regression (LR), Naive Bayes (NB), Support Vector Machine (SVM), Random Forest (RF), Neural Network (NN) and Decision Tree (DT) in order to demonstrate its utility. We have experimentally exhibited the exceptional empirical effectiveness and biological applicability of our suggested hybrid model for cancer diagnostics and medication development. © 2023, CC BY.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

The Effect of Human v/s Synthetic Test Data and Round-tripping on Assessment of Sentiment Analysis Systems for Bias

arXiv

引用

arXiv 2024年

作者： Lakkaraju, Kausik Gupta, Aniket Srivastava, Biplav Valtorta, Marco Wu, Dezhi AI Institute University of South Carolina ColumbiaSC United States Dept. of Computer Science Netaji Subhas University of Technology Delhi India Dept. of Computer Science and Engineering University of South Carolina ColumbiaSC United States Dept. of Integrated Information Technology University of South Carolina ColumbiaSC United States

Sentiment Analysis Systems (SASs) are data-driven Artificial Intelligence (AI) systems that output polarity and emotional intensity when given a piece of text as input. Like other AIs, SASs are also known to have unstable behavior when subjected to changes in data which can make them problematic to trust out of concerns like bias when AI works with humans and data has protected attributes like gender, race, and age. Recently, an approach was introduced to assess SASs in a blackbox setting without training data or code, and rating them for bias using synthetic English data. We augment it by introducing two human-generated chatbot datasets and also considering a round-trip setting of translating the data from one language to the same through an intermediate language. We find that these settings show SASs performance in a more realistic light. Specifically, we find that rating SASs on the chatbot data showed more bias compared to the synthetic data, and round-tripping using Spanish and Danish as intermediate languages reduces the bias (up to 68% reduction) in human-generated data while, in synthetic data, it takes a surprising turn by increasing the bias! Our findings will help researchers and practitioners refine their SAS testing strategies and foster trust as SASs are considered part of more mission-critical applications for global use. © 2024, CC BY-NC-ND.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Precision Agriculture: Crop Image Segmentation and Loss Evaluation through Drone Surveillance

Precision Agriculture: Crop Image Segmentation and Loss Eval...

引用

International Conference on Secure Cyber Computing and Communication (ICSCCC)

作者： Pawan Kumar Patidar Deepak Singh Tomar R. K. Pateriya Yogesh Kumar Sharma Dept. of Computer Science Engineering Maulana Azad National Institute of Technology Bhopal INDIA

Agriculture is the most important source of livelihood. Crop segmentation has become an important role in precision agriculture which helps farmers to make decisions about crop damage and its production. However, it's a challenging task to achieve precision in the agriculture field. Drone Surveillance helps to achieve that crop yield assessment, crop damage, crop health, and other parameters. This paper focuses on image segmentation of crops, classified into categories like sparse and dense crops with the multitemporal data image taken by Drone. This model proposed and studied shows the loss percentage in crop identification by image segmentation process, it helps farmers to get good compensation for crops to survey through Drone (UAV) techniques. A detailed analysis with outcome of thisis explained further.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Implementation of a Word Prediction Model Using LSTM Neural Network with Optimization Strategies

Implementation of a Word Prediction Model Using LSTM Neural ...

引用

Multi-Agent Systems for Collaborative Intelligence (ICMSCI), International Conference on

作者： Jagadevi P Yeresime Suresh Dept. of Computer Science & Engg. Ballari Institute of Technology and Management Ballari Visvesvaraya Technological University Belagavi Karnataka India Dept. of CSE - Artificial Intelligence Ballari Institute of Technology and Management Ballari

ISBN: (数字)9798331509828

ISBN: (纸本)9798331509835

The development of word predictive models has significantly advanced with the integration of deep learning techniques, particularly in the domain of natural language processing (NLP). Early models like n-grams and basic Recurrent Neural Networks (RNNs), laid the foundation for sequence based text generation by predicting the next word based on a fixed number of preceding words. However such approaches frequently struggled with capturing long range dependencies due to their limited context window and inability to maintain memory over extended sequences. This paper presents the implementation of word prediction model using Long Short-Term Memory (LSTM) neural networks that predicts the next word in a given sequence. The proposed model is trained on a diverse dataset and evaluating performance through several test cases. The model incorporates several optimization strategies to enhance its performance regarding scalability, fault tolerance, and resource allocation. Experimental results demonstrate the effectiveness of these optimizations in improving prediction accuracy and computational efficiency.

关键词： Fault tolerance Accuracy Computational modeling Scalability Fault tolerant systems Predictive models Transformers Natural language processing Optimization Long short term memory

来源：评论

学校读者我要写书评

暂无评论

Physical-Design Aware Scan Cell Reordering for Low Power Testing

Physical-Design Aware Scan Cell Reordering for Low Power Tes...

引用

East-West Design & Test Symposium (EWDTS)

作者： Rohit Badjatya Vinay Rahul Kumar Jaynarayan T Tudu Masahiro Fujita Virendra Singh Dept. of Electrical Engineering Indian Institute of Technology Bombay Mumbai India Dept. of Computer Science Engineering Indian Institute of Technology Tirupati India Tirupati India Dept. of Electrical Engineering University of Tokyo Tokyo Japan

ISBN: (数字)9798331515768

ISBN: (纸本)9798331515775

Scan chain architecture is widely employed in modern VLSI design for test applications. However, it often leads to high power consumption during testing. The architecture experiences elevated simultaneous switching activity due to non-functional input patterns applied during test time, resulting in significant peak power in the circuits. This increased power consumption can cause IR drop issues, negatively impacting the chip's yield. This paper proposes an innovative approach to reduce both peak power and routing length by effectively reordering the scan chains. Experimental results show that the proposed approach is effective for the ISCAS'89 and ITC'99 benchmark circuits and this method achieves lower peak power and routing length compared to the state of the art.

关键词： Power demand Layout computer architecture Benchmark testing Very large scale integration Routing Vectors Timing Signal integrity Switching circuits

来源：评论

学校读者我要写书评

暂无评论

Penalty-Enabled Serverless Architecture for Cloud-based Startup Solutions

Penalty-Enabled Serverless Architecture for Cloud-based Star...

引用

International Conference on Inventive Computation Technologies (ICICT)

作者： Shajulin Benedict Rubiya Subair Tanya Gupta Vedanta S.P. Dept. of Computer Science and Engineering Indian Institute of Information Technology Kottayam Kerala India

The step to the success of startups is through overcoming competitors by adopting software innovations that improve businesses. Serverless computing model, recently, has intrigued a sizable number of startup professionals belonging to various sectors, including financial or IoT-enabled application developers. One of the main flaws is its heavy dependency on cloud providers, which can still result in hefty pricing to startups and stalling functions in applications. This article proposes a penaltyenabled serverless architecture for startups. The architecture can boost the economy of startups and can analyze the serverlessoriented cost-saving options in applications. The penalty-oriented approach could enable cloud architects, developers, and startups, to rethink the utilization of serverless functions; to gleam of with future innovations.

关键词：

来源：评论

学校读者我要写书评

暂无评论

3D AttU-NET for Brain Tumor Segmentation with a Novel Loss Function

3D AttU-NET for Brain Tumor Segmentation with a Novel Loss F...

引用

International Conference on Information Systems and computer Networks (ISCON)

作者： Rahul Roy B. Annappa Shubham Dodia Dept. of Computer Science and Engineering National Institute of Technology Karnataka Surathkal Mangalore India

In the United States of America (USA), every year 150,000 patients are registered with a secondary brain tumor that is not generated in the brain. This necessitates the need for early brain tumor detection, which in turn will help patients to live longer. For clinical evaluation and treatment, precise segmentation of brain tumors in MRI images is required. This process can be aided by machine learning and efficient image processing, but manual imaging can be time-consuming. In this study, we aim to develop an 3D automated segmentation algorithm with a novel loss function. A 3D attention UNET CNN model was trained using the novel loss function, which was calculated by taking the weighted average of dice loss and focal loss to overcome the class imbalance. Results show the enhancement in the segmentation performance of attention UNET model with an average increase of 5% in the Dice coefficient for all three classes. However, the model’s performance was not as strong for enhanced and core tumors. Further research may be needed to optimize performance in these areas.

关键词： Solid modeling Image segmentation Three-dimensional displays Machine learning algorithms Magnetic resonance imaging Manuals Brain modeling

来源：评论

学校读者我要写书评

暂无评论

Advancing Healthcare with GPGPU and Energy Efficient NoC Architectures: A Comprehensive Survey

Advancing Healthcare with GPGPU and Energy Efficient NoC Arc...

引用

Ambient Intelligence in Health Care (ICAIHC), International Conference on

作者： Panthagani Vijaya Babu Sonal Yadav Vijay Holimath Dept. of Computer Science & Engineering ViBram Silicon Hub National Institute of Technology Raipur India VividSparks IT Solutions Bangalore India

ISBN: (数字)9798331530259

ISBN: (纸本)9798331530266

General-purpose graphics Processing Units (GPGPU) have emerged as a transformative technology in healthcare and medical fields, harnessing their powerful parallel processing capabilities to handle complex computational challenges. Initially developed for graphics rendering, GPUs are now employed in high-performance tasks such as medical imaging, genomic analysis, drug discovery, and real-time patient monitoring, where vast data volumes and intensive computations are prevalent. GPGPU technology enhances processing efficiency, reduces latency, and supports faster, more accurate diagnostics and treatment planning. Complementing this advancement, Networks-On-Chip (NoC) designs, introduced in the early 2000s, have become a standard communication backbone for high-end CPUs and Systems-on-Chip (SoCs). Their low communication latency, high throughput, and energy efficiency make NoCs ideal for addressing the growing demands of GPU-based systems. However, achieving these performance objectives requires minimizing power dissipation, energy consumption, and costs. This research provides a comprehensive survey of NoC design models for multi-GPU systems, focusing on their role in energy-efficient and scalable architectures. It also highlights the impact of GPGPU in revolutionizing healthcare by meeting modern medical applications' computational and efficiency requirements.

关键词： Surveys Scalability Memory management graphics processing units Medical services Throughput Energy efficiency Real-time systems Computational efficiency Standards

来源：评论

学校读者我要写书评

暂无评论

Towards Adoption of Green Blockchain with Emphasis on Blockchain Type, Consensus Protocols, Data Sharding and Smart Contracts

Towards Adoption of Green Blockchain with Emphasis on Blockc...

引用

Blockchain and Distributed Systems Security (ICBDS), IEEE International Conference on

作者： Shraddha Phansalkar Deepa Mishra Neha Chaube Rahul Sonkamble Dept. of Computer Science & Engineering MIT Art Design and Technology University Pune Maharashtra

Blockchain and related Distributed Ledger Technologies (DLT) are anticipated to transform a the world of web from a centralised document-sharing platform to a comprehensive decentralised platform that facilitates the exchange of digital currency and supports autonomous management of digital assets. The central server is susceptible to attacks, distrust and collusion. If the web can assure reliable, safe, and responsible updates among independent participants without the need for a centralized server, the perception of a decentralised web can be re-instantiated. One of the essential technologies required to restore the openness of the internet while maintaining its security is distributed ledger technology (DLT). DLTs may now totally handle business and legal transactions online, creating a more trustworthy and accountable environment. Blockchain technology marks a major breakthrough by removing the need for a centralised trusted authority in a widely distributed network. Instead, a consensus must be reached among several sources of trust, based on an algorithm, that this transaction may be believed to be legitimate. The consensus algorithms in blockchain technology offer an immutable and permanent record of a transaction that is immutable, trustworthy and secured. Consensus algorithms are however energy consuming because of their computation heavy nature. This has been a biggest inhibition towards blockchain adoption. The energy needs for committing a blockchain transaction is also governed by whether it is a public/permissioned blockchain, its consensus algorithm, onchain-offchain data and the code complexity of smart contracts. The paper presents a state-of art evaluation of the current blockchain platforms and cryptos and evaluate them with the energy consumptions. Additionally, it also proposes a framework architecture of a green blockchain application. A green blockchain refers to the implementation of environmentally sustainable algorithms, tools and platforms

关键词：

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：