检索结果-内蒙古大学图书馆

Safeguarding Ethical AI: Detecting Potentially Sensitive data Re-Identification and Generation of Misleading or Abusive Content from Quantized Large Language Models

Safeguarding Ethical AI: Detecting Potentially Sensitive Dat...

引用

17th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2024

作者： Kollapally, Navya Martin Geller, James Department of Computer Science New Jersey Institute of Technology Newark United States Department of Data Science New Jersey Institute of Technology Newark United States

Research on privacy-preserving Machine Learning (ML) is essential to prevent the re-identification of health data ensuring the confidentiality and security of sensitive patient information. In this era of unprecedented usage of large language models (LLMs), LLMs carry inherent risks when applied to sensitive data, especially as LLMs are trained on trillions of words from the internet, without a global standard for data selection. The lack of standardization in training LLMs poses a significant risk in the field of health informatics, potentially resulting in the inadvertent release of sensitive information, despite the availability of context-aware redaction of sensitive information. The research goal of this paper is to determine whether sensitive information could be re-identified from electronic health records during Natural Language Processing (NLP) tasks such as text classification without using any dedicated re-identification techniques. We performed zero and 8-shot learning with the quantized LLM models FLAN, Llama2, Mistral, and Vicuna for classifying social context data extracted from MIMIC-III. In this text classification task, our focus was on detecting potential sensitive data re-identification and the generation of misleading or abusive content during the fine-tuning and prompting stages of the process, along with evaluating the performance of the classification. © 2024 by SCITEPRESS–science and Technology Publications, Lda.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

DFA-GNN: Forward Learning of Graph Neural Networks by Direct Feedback Alignment 38

DFA-GNN: Forward Learning of Graph Neural Networks by Direct...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Zhao, Gongpei Wang, Tao Lang, Congyan Jin, Yi Li, Yidong Ling, Haibin Key Laboratory of Big Data & Artificial Intelligence in Transportation Ministry of Education China School of Computer Science & Technology Beijing Jiaotong University China Department of Computer Science Stony Brook University United States

Graph neural networks (GNNs) are recognized for their strong performance across various applications, with the backpropagation (BP) algorithm playing a central role in the development of most GNN models. However, despite its effectiveness, BP has limitations that challenge its biological plausibility and affect the efficiency, scalability and parallelism of training neural networks for graph-based tasks. While several non-backpropagation (non-BP) training algorithms, such as the direct feedback alignment (DFA), have been successfully applied to fully-connected and convolutional network components for handling Euclidean data, directly adapting these non-BP frameworks to manage non-Euclidean graph data in GNN models presents significant challenges. These challenges primarily arise from the violation of the independent and identically distributed (i.i.d.) assumption in graph data and the difficulty in accessing prediction errors for all samples (nodes) within the graph. To overcome these obstacles, in this paper we propose DFA-GNN, a novel forward learning framework tailored for GNNs with a case study of semi-supervised learning. The proposed method breaks the limitations of BP by using a dedicated forward training mechanism. Specifically, DFA-GNN extends the principles of DFA to adapt to graph data and unique architecture of GNNs, which incorporates the information of graph topology into the feedback links to accommodate the non-Euclidean characteristics of graph data. Additionally, for semi-supervised graph learning tasks, we developed a pseudo error generator that spreads residual errors from training data to create a pseudo error for each unlabeled node. These pseudo errors are then utilized to train GNNs using DFA. Extensive experiments on 10 public benchmarks reveal that our learning framework outperforms not only previous non-BP methods but also the standard BP methods, and it exhibits excellent robustness against various types of noise and attacks. © 2024 Neura

关键词：

来源：评论

学校读者我要写书评

暂无评论

Real-Time Road Damage Detection and Infrastructure Evaluation Leveraging Unmanned Aerial Vehicles and Tiny Machine Learning

引用

IEEE Internet of Things Journal 2024年第12期11卷 21347-21358页

作者： Khan, Muhammad Waseem Obaidat, Mohammad S. Mahmood, Khalid Batool, Dania Badar, Hafiz Muhammad Sanaullah Aamir, Muhammad Gao, Wu Comsats University Islamabad Sahiwal Campus Department Of Computer Science Sahiwal57000 Pakistan Superior University Department Of Computer Science Lahore54000 Pakistan The University Of Jordan King Abdullah Ii School Of Information Technology Amman11942 Jordan University Of Science And Technology Beijing School Of Communication And Computing Engineering Beijing100083 China Amity University Amity School Of Engineering And Technology Uttar Pradesh Noida201301 India National Yunlin University Of Science And Technology Graduate School Of Intelligent Data Science Douliu64002 Taiwan University Of Palermo Department Of Engineering Palermo90133 Italy Emerson University Department Of Emerging Computing Technologies Multan60000 Pakistan Northwestern Polytechnical University Center For Bio-Inspired System-on-Chip School Of Computer Science Xi'an710060 China

Road damage detection (RDD) through computer vision and deep learning techniques can ensure the safety of vehicles and humans on the roads. Integrating unmanned aerial vehicles (UAVs) in RDD and infrastructure evaluation (IE) has also emerged as a key enabler, contributing significantly to data acquisition and real-time monitoring of road damages, such as potholes, cracks, and surface anomalies, facilitating proactive maintenance and improved road conditions. These UAVs are low-powered and resource-constrained devices that work autonomously to perform pattern detection and decision making leveraging tiny machine learning (Tiny ML) algorithms. These Tiny ML algorithms are designed to run on edge devices, IoT devices, UAVs, etc. In this study, the RDD2022 data set collected using UAVs and dashboard cameras of vehicles was utilized to train pure and mixed models that exhibit class instance imbalance in certain classes which is addressed by implementing data augmentation as a regularization technique. State-of-the-art two-stage detectors: Faster R-CNN ResNet101 and one-stage detectors: SSD MobileNet V1 FPN, YOLOv5, and Efficientdet D1 are employed. The results indicate that the two-stage detector achieved an impressive mean average precision (mAP) of 88.49% overall and 96.62% for focused classes. Notably, the state-of-the-art Efficientdet D1 approach achieved a competitive mAP of 86.47% overall and 95.12% for focused classes, with significantly lower computational cost. These findings highlight the potential of advanced object detection techniques, particularly Efficientdet D1, to enhance the accuracy and efficiency of RDD systems, thereby improving passenger safety and overall performance. © 2014 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Combating deepfakes: a comprehensive multilayer deepfake video detection framework

引用

Multimedia Tools and Applications 2024年第38期83卷 85619-85636页

作者： Rathoure, Nikhil Pateriya, R.K. Bharot, Nitesh Verma, Priyanka Department of Computer Science and Engineering Maulana Azad National Institute of Technology Mata Mandir Bhopal462003 India Data Science Institute University of Galway University road GalwayH91TK33 Ireland Department of Electronics and Computer Engineering University of Limerick Castletroy LimerickV94 T9PX Ireland

Deepfakes represent a class of synthetic media crafted with the aid of advanced deep learning techniques that exhibit an unparalleled degree of authenticity. The rapid advancement in Artificial Intelligence (AI) has contributed to an increase in the prevalence of deepfakes on the internet, consequently amplifying the spread of misinformation among the public. Consequently, the detection of deepfakes has become a pressing concern. In this context, we put forth a comprehensive framework for deepfake video detection, which is built upon three distinct layers. The first layer, termed as the RGB features extraction layer, is designed to identify potential signs of forgery within the spatial domain of analogous video frames. The second layer, known as the GAN features extraction layer, focuses on the extraction of forgery fingerprints in the high-frequency region. This layer is specifically engineered to detect the fingerprints left by the Generative Adversarial Network (GAN) process in fake videos and the traces of the imaging process in genuine videos. The third and final layer, referred to as the facial region intra-frame inconsistency feature extraction layer, is dedicated to uncovering the anomalies associated with the manipulation process. This is achieved by extracting features from both the inner and outer regions of the manipulated portion of a frame. The extensive experimental evaluations have underscored the superior performance of proposed approach in comparison to existing state-of-the-art methods. © The Author(s) 2024.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Advancements in Environmental data Analysis for Climate-Resilient Agriculture Using Remote Sensing and Deep Learning

引用

Remote Sensing in Earth Systems sciences 2024年第4期7卷 670-680页

作者： Nirmal, Khyati R. Reddy, Chilukala Mahender P, Jose Tulasi, Rajesh Maheswari, G. Uma Sapthami, I. SNJB’S Late Sau KBJ CoE Maharashtra Nashik India Department of Artificial Intelligence and Data Science GITAM Deemed to Be University Hyderabad India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R & amp D Institute of Science and Technology Chennai Avadi India Dept. of Computer Science and Engineering Koneru Lakshmaiah Education Foundation A.P Vaddeswaram Guntur Dist. India Department of Computer Science and Engineering RMK College of Engineering and Technology Chennai Pudhuvoyal India Department of Computer Science and Engineering MLR Institute of Technology Hyderabad India

Weather significantly influences agricultural productivity. Plant biotic and abiotic stressors are primarily induced by climate change, resulting in a detrimental effect on worldwide agricultural productivity. These two components are closely interconnected. This paper presents an innovative approach in smart agriculture for climate change combining remote sensing and a deep learning algorithm. The input is obtained as a multispectral environmental image and then subjected to noise reduction and normalisation processing. The image has been retrieved using a primary convolutional component with a stacked encoder model. The retrieved features are identified using ResNet graph reinforcement neural networks. The resulting classification output displays environmental imagery depicting climate variations. The agriculture sector has been analysed based on the classified climate analysis. The experimental results have been conducted on diverse farm datasets related to climate change, evaluating the detection accuracy, recall, mean average precision, normalized correlation, and F-1 score. The proposed method achieved a detection accuracy of 98%, a normalised correlation of 95%, a mean average precision of 92%, a recall of 97%, and an F-MEASURE of 94%. Machine learning can assist in monitoring and predicting the impact of climate change on food security, as indicated by the findings. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Climate change Deep learning model Remote sensing Smart agriculture Stacked encoder model

来源：评论

学校读者我要写书评

暂无评论

Hyper parameter Optimization Using Genetic Algorithm for Extracting Social Determinants of Health Text

Hyper parameter Optimization Using Genetic Algorithm for Ext...

引用

17th International Joint Conference on Biomedical Engineering Systems and Technologies, BIOSTEC 2024

作者： Martin Kollapally, Navya Geller, James Department of Computer Science New Jersey Institute of Technology Newark United States Department of Data Science New Jersey Institute of Technology Newark United States

Clinical factors account only for a small portion, about 10-30%, of the controllable factors that affect an individual's health outcomes. The remaining factors include where a person was born and raised, where he/she pursued their education, what their work and family environment is like, etc. These factors are collectively referred to as Social Determinants of Health (SDoH). Our research focuses on extracting sentences from clinical notes, using an SDoH ontology (called SOHO) to provide appropriate concepts. We utilize recent advancements in Deep Learning to optimize the hyper parameters of a Clinical BioBERT model for SDoHtext. A genetic algorithm-based hyper parameter tuning regimen improved with principles of simulated annealing was implemented to identify optimal hyper parameter settings. To implement a complete classifier, we pipelined Clinical Bio BERT with two subsequent linear layers and two dropout layers. The output predicts whether a text fragment describes an SDoH issue of the patient. The proposed model is compared with an existing optimization framework for both accuracy of identifying optimal parameters and execution time. © 2024 by SCITEPRESS – science and Technology Publications, Lda.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

Artificial Vision - Yolov8 Algorithm for Objects and Impediments Detection 2

Artificial Vision - Yolov8 Algorithm for Objects and Impedim...

引用

2nd IEEE International Conference on Emerging Research in Computational science, ICERCS 2024

作者： Lakshmi, S. Jothi Sasirekha, S.P. Mekala, N. Reshma, R. Nair, V. Indu Bhagya, S.G. Nehru Institute of Technology Department of Information Technology Coimbatore India Karpagam Academy of Higher Education Department of Computer Science and Engineering Coimbatore India Coimbatore India SNS College of Engineering Department of Artificial Intelligence and Data Science Coimbatore India Bapuji Institute of Engineering and Technology Department of Artificial Intelligence and Data Science Davangere India

ISBN: (纸本)9798331534950

This paper presents an object detection method using YOLO technique (You Only Look Once) based deep learning algorithm to help visually impaired people in their daily life. A Cobotic Spectacle is a cutting-edge product designed specifically for visually impaired individuals. The system seamlessly integrates face motion recognition, object identification and text detection functionalities using YOLO Algorithm. The real time object detection method uses YOLO and convolution neural networks (CNN) for single forward propagation from end to end. The proposed system utilizes YOLOv8, a state-of-the-art real-time object detection algorithm, which is trained on the Common Object in Context (COCO) dataset to accurately identify objects in front of a person. With a high degree of accuracy, the system then conveys the identified object information through a synthesized voice output, thereby enhancing visual perception and providing an immersive *** face motion recognition module promotes improved social interaction, while object identification increases environmental awareness. This integrated solution combines edge processing and cloud-based analysis for portability, real-time processing, and computational efficiency. A Cobotic Spectacle empowers visually impaired individuals, providing a state-of-the-art smart spectacle system that enhances visual perception, promotes independence, and improves their overall quality of life. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Ensuring Ethical, Transparent, and Auditable Use of Education data and Algorithms on AutoML 6

Ensuring Ethical, Transparent, and Auditable Use of Educatio...

引用

6th International Conference on Machine Learning and Natural Language Processing, MLNLP 2023

作者： Griep, Kylie Stone, Leah Edwards, Stephen Hill, Malik Cui, Jialin Liu, Chengyuan Jia, Qinjin Song, Yang Department of Computer Science University of North Carolina Wilmington United States Master of Data Science Program University of North Carolina Wilmington United States Department of Computer Science NC State University United States

ISBN: (纸本)9798400709241

Automated machine learning (AutoML) creates additional opportunities for less advanced users to build and test their own data mining models. Even though AutoML creates the models for the user, there is still technical knowledge and tools needed to evaluate those models, and due to the black-box nature of the machine learning models, problems can arise with regard to algorithmic biases and fairness. Such biases can escalate in future applications, necessitating a structured approach for fairness evaluation in AutoML. This involves defining fairness criteria, selecting appropriate metrics, assessing fairness across groups, and addressing biases. In the realm of educational data mining, where AutoML is prevalent, biases related to attributes like gender or race can lead to unethical outcomes. Since fairness metrics vary in definition and strength, and some may even contradict others, making fairness evaluation more complex. In this paper, ten fairness metrics were chosen, explored, and implemented on four AutoML tools, Vertex AI, AutoSklearn, AutoKeras, and PyCaret. We identified two open educational datasets and built both prediction and classification models on those AutoML frameworks. We report our work in evaluating different machine learning models created by AutoML and provide discussions about the challenges in evaluating fairness in those models and our effort to mitigate and resolve the problems of algorithmic bias in educational data mining. © 2023 ACM.

关键词： data mining

来源：评论

学校读者我要写书评

暂无评论

Prospector Heads: Generalized Feature Attribution for Large Models & data 41

Prospector Heads: Generalized Feature Attribution for Large ...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Machiraju, Gautam Derry, Alexander Desai, Arjun Guha, Neel Karimi, Amir-Hossein Zou, James Altman, Russ B. Ré, Christopher Mallick, Parag Department of Biomedical Data Science Stanford University United States Cartesia AI United States Department of Computer Science Stanford University United States Department of Electrical & Computer Engineering University of Waterloo Canada Department of Radiology Stanford University United States

Feature attribution, the ability to localize regions of the input data that are relevant for classification, is an important capability for ML models in scientific and biomedical domains. Current methods for feature attribution, which rely on "explaining" the predictions of end-to-end classifiers, suffer from imprecise feature localization and are inadequate for use with small sample sizes and high-dimensional datasets due to computational challenges. We introduce prospector heads, an efficient and interpretable alternative to explanation-based attribution methods that can be applied to any encoder and any data modality. Prospector heads generalize across modalities through experiments on sequences (text), images (pathology), and graphs (protein structures), outperforming baseline attribution methods by up to 26.3 points in mean localization AUPRC. We also demonstrate how prospector heads enable improved interpretation and discovery of class-specific patterns in input data. Through their high performance, flexibility, and generalizability, prospectors provide a framework for improving trust and transparency for ML models in complex domains. Copyright 2024 by the author(s)

关键词： data assimilation

来源：评论

学校读者我要写书评

暂无评论

Classification Attack on IoT Devices with Machine Learning 9

Classification Attack on IoT Devices with Machine Learning

引用

9th IEEE International Conference on Computational Intelligence and Applications, ICCIA 2024

作者： Charoenwong, Nutcha Kosolsombat, Somkiat Ratanavilisagul, Chiabwoot College of Interdisciplinary Studies Thammasat University Data Science and Innovation Pathum Thani Thailand King Mongkut's University of Technology Faculty of Applied Science Department of Computer and Information Science North Bangkok Thailand

ISBN: (纸本)9798350352214

Internet of Things (IoT) devices are small, low-power devices used for detecting and processing data remotely through the internet. These devices have increasingly been integrated into our daily lives, both in the digital and physical world. However, this interconnectedness poses risks of being targeted by malware due to limited resources and connectivity. Machine learning (ML) techniques can be used to identify and prevent malware attacks on IoT devices. In this research report, we investigate the use of ML techniques for classifying malware in IoT devices. Our research includes the implementation and evaluation of decision tree, k-nearest neighbors (KNN), and XGBoost algorithms. Our evaluation metrics include accuracy, precision, recall, and the average of precision and recall. Our method is tested using a dataset of real-world malware attacks on IoT devices, and the XGBoost algorithm achieves the highest accuracy of 98.44%, demonstrating its effectiveness in detecting and classifying malware in IoT devices. © 2024 IEEE.

关键词： Cybersecurity

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：