检索结果-内蒙古大学图书馆

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Rashmi Adyapady, R. Annappa, B. Sagar, Prem Nmam Institute of Technology Department of Artificial Intelligence and Machine Learning Karnataka Nitte India National Institute of Technology Karnataka Discover Lab Department of Computer Science and Engineering Karnataka Mangalore India

ISBN: (纸本)9798350370249

Image inpainting is a domain in which researchers have shown considerable interest, and when it comes to deep learning techniques, realistic problems become interesting and challenging. In image inpainting, a corrupted facial image with missing holes or significant holes can be restored and compared to the original image to see if it is real or fake. In addition to fixing the texture of the image and getting the image's high-level abstract properties, it may also recover semantic images such as human faces. In the field of image-inpainting models, the Attention model with features learned through semantic approaches and progressive networks has become particularly popular. The proposed model introduces (i) Attention blocks in each decoder layer of U-Net architecture and (ii) a hybrid loss function leveraging both Mean Square Error (MSE) and Mean Absolute Error (MAE). The proposed Attention-based U-Net showed remarkable performance with SSIM and PSNR by 0.1067 and 13.63, respectively, compared to the previous approaches. © 2024 IEEE.

关键词： Distorted Images Image Inpainting Partial Convolution U-Net

来源：评论

学校读者我要写书评

暂无评论

Adversarial learning Based Knowledge Distillation on 3D Point Clouds

Adversarial Learning Based Knowledge Distillation on 3D Poin...

引用

2025 IEEE/CVF Winter Conference on Applications of computer Vision, WACV 2025

作者： Sanjay, S.J. Akash, J. Rajan, Sreehari Shajahan, Dimple A Sharma, Charu Iiit Hyderabad Machine Learning Lab India Tkm College of Engineering Department of Computer Science Kollam India

ISBN: (纸本)9798331510831

The significant improvements in point cloud representation learning have increased its applicability in many real-life applications, resulting in the need for lightweight, better-performing models. One widely proposed efficient method is knowledge distillation, where a lightweight model uses knowledge from large models. Very few works exist on distilling the knowledge for point clouds. Most of the work focuses on cross-modal-based approaches that make the method expensive to train. This paper proposes PointKAD, an adversarial knowledge distillation framework for point cloud-based tasks. PointKAD includes adversarial feature distillation and response distillation with the help of discriminators to extract and distill the representation of feature maps and logits. We conduct extensive experimental studies on both synthetic (ModelNet40) and real (ScanObjectNN) datasets to show that PointKAD achieves state-of-the-art results compared to the existing knowledge distillation methods for point cloud classification. Additionally, we present results on the part segmentation task, highlighting the efficacy of the PointKAD framework. Our experiments further reveal that PointKAD is capable of transferring knowledge across different tasks and datasets, show-casing its versatility. Furthermore, we demonstrate that PointKAD can be applied to a cross-modal training setup, achieving competitive performance with cross-modal-based point cloud methods for classification. © 2025 IEEE.

关键词： Modal analysis

来源：评论

学校读者我要写书评

暂无评论

Analyzing the Feasibility of Bert Model for Toxicity Analysis of Text 6th

Analyzing the Feasibility of Bert Model for Toxicity Analysi...

引用

6th International Conference on Innovative Computing and Communication, ICICC 2023

作者： Chakraverty, Yuvraj Kaintura, Aman Kumar, Bharat Khanna, Ashish Sharma, Moolchand Pareek, Piyush Kumar Department of Computer Science and Engineering Maharaja Agrasen Institute of Technology New Delhi India Department of Artificial Intelligence and Machine Learning and IPR Cell Nitte Meenakshi Institute of Technology Bengaluru India

ISBN: (纸本)9789819940707

Online comments can often be toxic, offensive, and harmful to individuals and communities. In recent years, there has been a growing need to automatically identify and mitigate these toxic comments. For this problem, NLP models are often used to identify such toxicity and harshness but each model has its own efficiency and performance limitations. In this paper, we propose the use of the bidirectional encoder representations from transformers (BERT) algorithm for toxicity classification of online comments. BERT is a state-of-the-art natural language processing model developed by Google in 2018 that has shown strong results on a variety of tasks. In this paper, we used the BERT algorithm for toxicity classification and evaluated its performance on a real world dataset and performed comparative analysis with conventional NLP models, logistic regression (TF-IDF) over which BERT showed an improvement of 6.9% in accuracy, 26.1% in f1-score, 21.5% in ROC score;logistic regression (BOW) over which BERT showed an improvement of 9.1% in accuracy, 70.6% in f1-score, 39.8% in ROC score;multinomialNB (BOW) over which BERT showed an improvement of 9.2% in accuracy, 25.9% in f1-score, 10.6% in ROC score. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Understanding Finetuning for Factual Knowledge Extraction 41

Understanding Finetuning for Factual Knowledge Extraction

引用

41st International Conference on machine learning, ICML 2024

作者： Ghosal, Gaurav Hashimoto, Tatsunori Raghunathan, Aditi Department of Machine Learning Carnegie Mellon University Pittsburgh United States Department of Computer Science Stanford University Stanford United States

In this work, we study the impact of QA finetuning data on downstream factuality. We show that fine-tuning on lesser-known facts that are poorly stored during pretraining yields significantly worse factuality than fine-tuning on well-known facts, even when all facts are seen during pretraining. We prove this phenomenon theoretically, showing that training on lesser-known facts can lead the model to ignore subject entity names and instead output a generic plausible response even when the relevant factual knowledge is encoded in the model. On three question answering benchmarks (PopQA, Entity Questions, and MMLU) and two language models (Llama-2-7B and Mistral-7B), we find that (i) finetuning on a completely factual but lesser-known subset of the data deteriorates downstream factuality (5-10%) and (ii) finetuning on a subset of better-known examples matches or outperforms finetuning on the entire dataset. Ultimately, our results shed light on the interaction between pretrained knowledge and finetuning data and demonstrate the importance of taking into account how facts are stored in the pretrained model when fine-tuning for knowledge-intensive tasks. Copyright 2024 by the author(s)

关键词： Question answering

来源：评论

学校读者我要写书评

暂无评论

DBNLP: detecting bias in natural language processing system for India-centric languages

引用

International Journal of Information Technology (Singapore) 2025年 1-16页

作者： T. G, Keerthan Kumar Mendke, Saish Parihar, Rohit Mayya, Samarth Venkatesh, Spoorthy Koolagudi, Shashidhar G. Department of Information Science and Engineering Siddaganga Institute of Technology Karnataka Tumkur 572102 India Department of Computer Science and Engineering National Institute of Technology Karnataka Surathkal Karnataka Mangalore 575025 India Department of Artificial Intelligence and Machine Learning CSPIT CHARUSAT Gujarat Anand 388421 India

Natural language processing (NLP) is gaining widespread interest and seeing advancements rapidly due to its attractive and exhilarating applications. NLP models are being developed in search engines for real-world scenarios such as language translation, sentiment analysis, chat-bots such as ChatGPT, and auto-completion. These models are trained on a vast corpus of online data, exposing them to harmful biases and stereotypes towards various communities. The models learn these biases, making harmful and undesirable predictions about particular genders, religions, races, and professions. Biases in NLP systems can perpetuate societal biases and discrimination, leading to unfair and unequal treatment of individuals or groups. It is crucial to identify these biases, which will help mitigate them. Most of the literary works in this area have been primarily Western-centric, focusing on the English language, making it tough to use them for Indian models and languages. In this work, we propose a model called Detecting Bias in Natural Language Processing System for India-Centric Languages (DBNLP), which aims to identify the biases relevant to the Indian context present in the text-based language models, particularly for the English and Hindi languages. The DBNLP presents three techniques for bias identification based on (1) a Context Association Test (CAT), (2) a template-based perturbation technique for various co-domain associations, and (3) a co-occurrence count-based corpus analysis technique. Further, this work showcases how India-centric models such as IndicBERT, MuRIL, and datasets such as IndicCorp are biased toward various demographic categories. Detecting bias in natural language processing systems for India-centric languages is essential to creating fair, diverse, and inclusive models that benefit society. © Bharati Vidyapeeth's institute of computer Applications and Management 2025.

关键词： Diverse and Inclusive models Harmful biases and stereotypes India-centric models Indian context Natural language processing

来源：评论

学校读者我要写书评

暂无评论

Hybrid Deep learning Cloud Intrusion Detection 3

Hybrid Deep Learning Cloud Intrusion Detection

引用

3rd IEEE International Conference on Intelligent Techniques in Control, Optimization and Signal Processing, INCOS 2024

作者： Karnan, Lokanayaki Mahalakshmi, S.B. Thirunavukkarasu, V. Christ University Department of Computer Science Karnataka Bangalore560029 India Coimbatore Institute of Technology Department of Artificial Intelligence & Machine Learning Tamil Nadu Coimbatore641014 India

ISBN: (纸本)9798350361186

The scalability and flexibility that cloud computing provides, organisations can readily adapt their resources to meet demand without having to make significant upfront expenditures in hardware infrastructure. Three main types of computing services are provided to people worldwide via the Internet. Increased performance and resource access are two benefits that come with using cloud computing, but there is also an increased chance of attack. As a result of this research, intrusion detection systems that can process massive amounts of data packets, analyse them, and produce reports using knowledge and behaviour analysis were created. Convolution Neural Network Algorithm encrypts data as it's being transmitted end-to-end and is stored in the cloud, providing an extra degree of security. Data protection in the cloud is improved by intrusion detection. This study uses a model to show how data is encrypted and decrypted, of an algorithm and describes the defences against attacks. When assessing the performance of the suggested system, it's critical to consider the time and memory needed to encrypt and decrypt big text files. Additionally, the security of the cloud has been investigated and contrasted with various encoding techniques now in use. © 2024 IEEE.

关键词： Intrusion detection

来源：评论

学校读者我要写书评

暂无评论

Brain Tumor Detection using YOLO v7 and YOLO v8

Brain Tumor Detection using YOLO v7 and YOLO v8

引用

2024 IEEE International Conference on Modeling, Simulation and Intelligent Computing, MoSICom 2024

作者： Vaghela, Rohan Sarda, Jigar Thakkar, Amit Vaishnani, Darsh Khokhariya, Avani Mehta, Pooja Charotar University of Science & Technology Department of Computer Science & Engineering Anand India Charotar University of Science & Technology Department of Artificial Intelligence & Machine Learning Anand India

ISBN: (纸本)9798331533311

The architecture of the brain is incredibly complex, with multiple areas managing various parts of the nervous system. An adult's health is greatly at risk from brain tumors because of their ability to severely impair organ function or possibly cause death due to their rapid growth of abnormal brain cells. Brain tumors can develop in any area of the skull or brain, including the protective layer around the brain and many other areas. Since these tumors can have a wide range of shapes, sizes, and locations, magnetic resonance imaging (MRI) is an essential tool for locating malignant tumors. MRIs are examined manually in order to diagnose brain tumors. However, manually identifying brain tumors is a challenging and time-consuming task that may result in errors. As a result, computer-aided brain tumor detection has been considered a crucial area of study in biomedical engineering. However, because tumor sizes and shapes vary enormously throughout subjects, developing a reliable brain tumor detector remains difficult. Thus, for the purpose of accurately identifying brain tumors in MRI scans, two popular deep learning detection algorithms - YOLO v7 and YOLO v8 - were tested and compared in this study. From the results, YOLO v8 model has achieved highest validation accuracy of 96.50% with average precision and recall value of 95.9% and 93.3% respectively, while YOLO v7 has validation accuracy of 87.6% with average precision and recall value of 81.9% and 85.4% respectively. © 2024 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

PAC-Bayes-Chernoff bounds for unbounded losses 38

PAC-Bayes-Chernoff bounds for unbounded losses

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Casado, Ioar Ortega, Luis A. Pérez, Aritz Masegosa, Andrés R. Spain Machine Learning Group Computer Science Dept. EPS Universidad Autónoma de Madrid Spain Department of Computer Science Aalborg University Denmark

We introduce a new PAC-Bayes oracle bound for unbounded losses that extends Cramér-Chernoff bounds to the PAC-Bayesian setting. The proof technique relies on controlling the tails of certain random variables involving the Cramér transform of the loss. Our approach naturally leverages properties of Cramér-Chernoff bounds, such as exact optimization of the free parameter in many PAC-Bayes bounds. We highlight several applications of the main theorem. Firstly, we show that our bound recovers and generalizes previous results. Additionally, our approach allows working with richer assumptions that result in more informative and potentially tighter bounds. In this direction, we provide a general bound under a new model-dependent assumption from which we obtain bounds based on parameter norms and log-Sobolev inequalities. Notably, many of these bounds can be minimized to obtain distributions beyond the Gibbs posterior and provide novel theoretical coverage to existing regularization techniques. © 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

Identification of Infectious Disease Like COVID-19 Gene Biomarkers Using a Clear Artificial Intelligence Approach

引用

SN computer science 2025年第4期6卷 1-13页

作者： Dubey, Shivendra Verma, Dinesh Kumar Kumar, Mahesh Department of Artificial Intelligence and Machine Learning Manipal University Jaipur Rajasthan Jaipur India Department of Computer Science and Engineering Jaypee University of Engineering and Technology MP Guna India

This study offers an approach that uses SARS-COV-2 mNGS (meta-genomic next-generation sequencing) samples to apply XAI (explainable artificial intelligence) methodologies derived from the use of machine learning methods. The research's data set contains 15,979 expressions of genes from 234 infected persons, of whom 39.68% (93) had SARS-COV-2 that was positive and 60.29% (141) had SARS-COV-2 that was negatives. The SARS-COV-2 related gene was chosen using the LASSO (least absolute shrinkage and selection operator) technique. The class imbalanced issue has been resolved using the SVM-SMOTE (Support Vector machine - Synthetic Minority Oversampling Technique) approach. To identify SARS-COV-2 related biomarkers possibilities and increase the comprehensibility of the final approach, an explainable strategy that utilized SHAP (SHAPley Additive exPlanations) as well as LIME (local interpretable model-agnostic explanations) approaches were used. The XGBoost (92.98% accuracy) approach performs better for the identification of infectious diseases like COVID-19 than the LR (91.17% accuracy), SVM (87.68% accuracy), and RF (91.18% accuracy) models. The three most significant genes connected to COVID-19 by the SHAP method were FAM83A, LGR6, and IFI27. The LIME findings demonstrated that the likelihood of a positive class was particularly influenced by the elevated degree of IFI27 expression of genes. Infectious diseases like COVID-19 could be accurately predicted using the suggested framework (XGBoost). The findings demonstrate that SHAP and LIME, along with the exploit of ML methods, it could be clarify the biological marker prognosis for infectious disease and give health center a clear knowledge about how potential risk factors affect the forecasts made by the model. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： Biomarker Gene LIME SARS SHAP

来源：评论

学校读者我要写书评

暂无评论

Polyhouse Agriculture with AI: Strategies for Climate Control, Energy Efficiency and Yield Improvement

引用

SN computer science 2024年第8期5卷 1119页

作者： Mayuri, K.P. Kathavate, Sheela Niranjanamurthy, M. Department of Artificial Intelligence and Machine Learning BMS Institute of Technology and Management (Affiliated to VTU -Belgaum) Karnataka Bengaluru India Department of Information Science and Engineering BMS Institute of Technology and Management Karnataka Bengaluru India Department of Artificial Intelligence and Machine Learning BMS Institute of Technology and Management Karnataka Bengaluru India

Artificial Intelligence is applied to an extensive array of climate change responses, from data collection and classification to conservation decision-making and rule enforcement, leading to improved understanding and mitigation efforts. Greenhouse and polyhouse farming practices could significantly add to the carbon footprint, posing a challenge to be addressed if sustainable and environmentally friendly solutions are to be assured. This study describes the use of Artificial Intelligence in polyhouse agriculture to achieve better yields, save energy and advocate for sustainable farming. This study further elucidates the impact of AI-driven climate control systems on the quality of crops, resource use efficiency and productivity. The findings demonstrated the enormous potential of AI technology to improve plant growth, lower greenhouse gas emissions and increase sustainability in polyhouse agriculture. AI-driven climate control systems optimize the growing environment and reduce energy use with reduced emissions of GHGs, thus contributing to sustainable farming. However, this study also emphasizes the need for detailed research into the deployment and consequences of AI technology within polyhouse operations. The key bottlenecks are availability, security and regulatory compliance related to the generation of data. Therefore, need for a holistic approach to the design phase of AI-driven climate control systems in polyhouses. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： AI-driven climate control Crop yield Greenhouse gas emissions Polyhouse agriculture Resource efficiency Sustainable farming

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：