检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Qin, Jiuming Liu, Che Cheng, Sibo Guo, Yike Arcucci, Rossella Department of Computing Imperial College London United Kingdom Data Science Institute Imperial College London United Kingdom Department of Earth Science and Engineering Imperial College London United Kingdom Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong

Modern healthcare often utilises radiographic images alongside textual reports for diagnostics, encouraging the use of Vision-Language Self-Supervised Learning (VL-SSL) with large pre-trained models to learn versatile medical vision representations. However, most existing VL-SSL frameworks are trained end-to-end, which is computation-heavy and can lose vital prior information embedded in pre-trained encoders. To address both issues, we introduce the backbone-agnostic Adaptor framework, which preserves medical knowledge in pre-trained image and text encoders by keeping them frozen, and employs a lightweight Adaptor module for cross-modal learning. Experiments on medical image classification and segmentation tasks across three datasets reveal that our framework delivers competitive performance while cutting trainable parameters by over 90% compared to current pre-training approaches. Notably, when fine-tuned with just 1% of data, Adaptor outperforms several Transformer-based methods trained on full datasets in medical image segmentation. © 2024, CC BY.

关键词： Signal encoding

来源：评论

学校读者我要写书评

暂无评论

Convergence rates for the Adam optimizer

arXiv

引用

arXiv 2024年

作者： Dereich, Steffen Jentzen, Arnulf Institute for Mathematical Stochastics University of Münster Germany School of Data Science Shenzhen Research Institute of Big Data The Chinese University of Hong Kong Shenzhen [CUHK-Shenzhen China Applied Mathematics: Institute for Analysis and Numerics University of Münster Germany

Stochastic gradient descent (SGD) optimization methods are nowadays the method of choice for the training of deep neural networks (DNNs) in artificial intelligence systems. In practically relevant training problems, usually not the plain vanilla standard SGD method is the employed optimization scheme but instead suitably accelerated and adaptive SGD optimization methods are applied. As of today, maybe the most popular variant of such accelerated and adaptive SGD optimization methods is the famous Adam optimizer proposed by Kingma & Ba in 2014. Despite the popularity of the Adam optimizer in implementations, it remained an open problem of research to provide a convergence analysis for the Adam optimizer even in the situation of simple quadratic stochastic optimization problems where the objective function (the function one intends to minimize) is strongly convex. In this work we solve this problem by establishing optimal convergence rates for the Adam optimizer for a large class of stochastic optimization problems, in particular, covering simple quadratic stochastic optimization problems. The key ingredient of our convergence analysis is a new vector field function which we propose to refer to as the Adam vector field. This Adam vector field accurately describes the macroscopic behaviour of the Adam optimization process but differs from the negative gradient of the objective function (the function we intend to minimize) of the considered stochastic optimization problem. In particular, our convergence analysis reveals that the Adam optimizer does typically not converge to critical points of the objective function (zeros of the gradient of the objective function) of the considered optimization problem but converges with rates to zeros of this Adam vector field. Even though our convergence results are only formulated for the Adam optimizer, the arguments in our convergence analysis can also be applied to other related SGD optimization methods and open the door for a sys

关键词： Vectors

来源：评论

学校读者我要写书评

暂无评论

Spatial-temporal evolution analysis of Dongting Lake using multi-source remote sensing data 6

Spatial-temporal evolution analysis of Dongting Lake using m...

引用

6th International Conference on Geoscience and Remote Sensing Mapping, GRSM 2024

作者： Zheng, Yuanmao Fu, Mingzhe He, Qiuhua He, Yuanrong Huang, Jianwei Liang, Xueyu Liu, Xinyue School of Environmental Science and Engineering Xiamen University of Technology Xiamen361024 China Technology Innovation Center for Ecological Conservation and Restoration in Dongting Lake Basin Ministry of Natural Resources Changsha410007 China Big Data Institute of Digital Natural Disaster Monitoring in Fujian Xiamen University of Technology Xiamen361024 China

ISBN: (纸本)9781510688018

This study analyzes the Dongting Lake's spatiotemporal changes using long-term remote sensing data. The key findings include: (1) From 2005 to 2015, the lake's water area decreased by 61.12% during the wet season and increased by 50.38% in the arid season. (2) The centroid migration was most significant from 2005 to 2010 (8565.82 meters) due to water area fluctuations but decreased to 0.95 meters from 2010 to 2015. (3) Climate factors, particularly precipitation, had a moderate correlation with area changes (0.20 in the wet season, 0.31 in the dry season). The study also explores the impact of human activities on water area changes, providing valuable insights for sustainable development strategies. © 2025 SPIE.

关键词： Lakes

来源：评论

学校读者我要写书评

暂无评论

On the Element-Wise Representation and Reasoning in Zero-Shot Image Recognition: A Systematic Survey

arXiv

引用

arXiv 2024年

作者： Guo, Jingcai Rao, Zhijie Chen, Zhi Guo, Song Zhou, Jingren Tao, Dacheng Department of Computing The Hong Kong Polytechnic University Hong Kong School of Electrical Engineering and Computer Science The University of Queensland BrisbaneQLD4072 Australia Department of Computer Science and Engineering Hong Kong University of Science and Technology Hong Kong Alibaba Group Hangzhou311121 China College of Computing and Data Science Nanyang Technological University Singapore

Zero-shot image recognition (ZSIR) aims to recognize and reason in unseen domains by learning generalized knowledge from limited data in the seen domain. The gist of ZSIR is constructing a well-aligned mapping between the input visual space and the target semantic space, which is a bottom-up paradigm inspired by the process by which humans observe the world. In recent years, ZSIR has witnessed significant progress on a broad spectrum, from theory to algorithm design, as well as widespread applications. However, to the best of our knowledge, there remains a lack of a systematic review of ZSIR from an element-wise perspective, i.e., learning fine-grained elements of data and their inferential associations. To fill the gap, this paper thoroughly investigates recent advances in element-wise ZSIR and provides a sound basis for its future development. Concretely, we first integrate three basic ZSIR tasks, i.e., object recognition, compositional recognition, and foundation model-based open-world recognition, into a unified element-wise paradigm and provide a detailed taxonomy and analysis of the main approaches. Next, we summarize the benchmarks, covering technical implementations, standardized datasets, and some more details as a library [link]. Last, we sketch out related applications, discuss vital challenges, and suggest potential future directions. Copyright © 2024, The Authors. All rights reserved.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

TinyML Benchmark: Executing Fully Connected Neural Networks on Commodity Microcontrollers 7

TinyML Benchmark: Executing Fully Connected Neural Networks ...

引用

7th IEEE World Forum on Internet of Things, WF-IoT 2021

作者： Sudharsan, Bharath Salerno, Simone Nguyen, Duc-Duy Yahya, Muhammad Wahid, Abdul Yadav, Piyush Breslin, John G. Ali, Muhammad Intizar Data Science Institute Galway NUI Ireland Dublin City University School of Electronic Engineering Ireland

ISBN: (纸本)9781665444316

Recent advancements in the field of ultra-low-power machine learning (TinyML) promises to unlock an entirely new class of edge applications. However, continued progress is restrained by the lack of benchmarking Machine Learning (ML) models on TinyML hardware, which is fundamental to this field reaching maturity. In this paper, we designed 3 types of fully connected Neural Networks (NNs), trained each NN using 10 datasets (produces 30 NNs), and present the benchmark by reporting the onboard model performance on 7 popular MCU-boards (similar boards are used to design TinyML hardware). We open-sourced and made the complete benchmark results freely available online1 to enable the TinyML community researchers and developers to systematically compare, evaluate, and improve various aspects during the design phase of ML-powered IoT hardware. © 2021 IEEE.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

Effective Concurrency Testing for Go via Directional Primitive-Constrained Interleaving Exploration

Effective Concurrency Testing for Go via Directional Primiti...

引用

IEEE International Conference on Automated Software Engineering (ASE)

作者： Zongze Jiang Ming Wen Yixin Yang Chao Peng Ping Yang Hai Jin School of Cyber Science and Engineering Huazhong University of Science and Technology China Hubei Key Laboratory of Distributed System Security Services Computing Technology and System Lab Cluster and Grid Computing Lab Hubei Engineering Research Center on Big Data Security National Engineering Research Center for Big Data Technology and System ByteDance Beijing China School of Computer Science and Technology Huazhong University of Science and Technology China

The Go language (Go/Golang) has been attracting increasing attention from the industry over recent years due to its strong concurrency support and ease of deployment. This programming language encourages developers to use channel-based concurrency, which simplifies the development of concurrent programs. Unfortunately, it also introduces new concurrency problems that differ from those caused by the mechanism of shared memory concurrency. However, there are only few works that aim to detect such Go-specific concurrency issues. Even state-of-the-art testing tools will miss critical concurrent bugs that require fine-grained and effective interleaving exploration. This paper presents GoPie, a novel testing approach for detecting Go concurrency bugs through primitive-constrained interleaving exploration. GoPie utilizes execution histories to identify new interleavings instead of relying on exhaustive exploration or random scheduling. To evaluate its performance, we applied GoPie to existing benchmarks and large-scale open-source projects. Results show that GoPie can effectively explore concurrent interleavings and detect significantly more bugs in the benchmark. Furthermore, it uncovered 11 unique previously unknown concurrent bugs, and 9 of which have been confirmed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Advancements in Artificial Intelligence for Better Image Analysis, Recognition, and Interpretation in the Field of Image Processing 2

Advancements in Artificial Intelligence for Better Image Ana...

引用

2nd IEEE Global Conference on Communications and Information Technologies, GCCIT 2024

作者： Saravanan, V. Adaikkammai, A. Sasikala, G. Iyswariya, A. Babisha, A. Latha, M. Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Nano Electronics Materials and Sensors Tamilnadu Chennai India R.M.D. Engineering College Department of Computer Science and Business Systems India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Department of Ece Avadi Chennai India R.M.K. Engineering College Department of Ece Chennai India Panimalar Engineering College Department of Artificial Intelligence and Data Science India Srm Institute of Science and Technology Department of It India

ISBN: (纸本)9798350388916

The rapid advancement of artificial intelligence (AI) has brought about a significant revolution in image processing. This revolutionary technology has enabled the analysis, recognition, and interpretation of images in ways that were previously inconceivable. An investigation into the most recent developments in artificial intelligence techniques that have improved the efficiency and precision of image processing tasks is presented in this study. We survey the emergence of transformer architectures and the integration of deep learning models, namely Convolutional Neural Networks (CNNs), to explore how these technologies have transformed the understanding and application of images in several industries. An overview of existing image processing methods and the limits of those approaches is presented at the beginning of our research. This highlights the necessity of more advanced AI-driven solutions. After that, we examine the advancements that have been made in artificial intelligence, such as the incorporation of attention processes, the development of more effective CNN architectures, and the utilisation of generative adversarial networks (GANs) for the purpose of picture synthesis and augmentation. The purpose of this research is to investigate the impact that artificial intelligence has on particular applications of image processing, such as autonomous car navigation systems, facial recognition, medical imaging, and satellite images analysis. Specifically, we focus on the enhancements in diagnostic accuracy, environmental monitoring, security, and safety that arose as a result of the adoption of artificial intelligence in these sectors through the use of case studies. Furthermore, we discuss the issues that are associated with artificial intelligence in image processing. These challenges include the necessity for explainable AI models, the requirement for computational resources, and the protection of data privacy. A number of potential solutions to these problems

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Diabetic Retinopathy Screening with Sequential Deep Learning Models

Enhancing Diabetic Retinopathy Screening with Sequential Dee...

引用

International Conference on Image Information Processing

作者： Vankadari Mohith Kalpana Raja I. R. Oviya Department of Computer Science and Engineering (AIE) Amrita School of Computing Amrita Vishwa Vidyapeetham Chennai India Section for Biomedical Informatics and Data Science School of Medicine Yale University New Haven USA

ISBN: (数字)9798350371406

ISBN: (纸本)9798350371413

Diabetic retinopathy is a serious medical disorder that, if left untreated, can result in visual impairment or blindness. The precise and timely categorization of its severity is critical for appropriate medical management. This research describes an automated method for determining the severity of diabetic retinopathy using deep learning algorithms. The suggested methodology adopts a two-step process for identifying diabetic retinopathy and then categorizing its severity. In this technique, two pre-trained deep learning models, InceptionResNetV2 and DenseNet121, are used. For training and assessment purposes, the “APTOS 2019 Blindness Detection” dataset which consists of five severity classes is used as the foundation. To overcome class imbalance, data augmentation approaches are used to improve the model’s performance. The results of the experiment show that the strategy is effective, with an accuracy of 83.10 % and an F1 score of 82.77 %. This study contributes to the development of automated systems for assessing diabetic retinopathy, perhaps enhancing early detection and treatment.

关键词： Deep learning Training Diabetic retinopathy Visual impairment Transfer learning Blindness Information processing

来源：评论

学校读者我要写书评

暂无评论

Accurate Generated Text Detection Based on Deep Layer-wise Relevance Propagation

Accurate Generated Text Detection Based on Deep Layer-wise R...

引用

IEEE International Conference on Big data Analysis (ICBDA)

作者： Mengjie Guo Limin Liu Meicheng Guo Siyuan Liu Zhiwei Xu Chinese Academy of Sciences Institute of Computing Technology Beijing China College of Data Science and Application Inner Mongolia University of Technology Hohhot China

Text generation models (TGMs) excel in producing text that matches the style of human language reasonably well. Such TGMs (e.g., GAN-based text generation model) can be misused for fake news and fake product reviews that can look authentic and fool humans. To distinguish the generated texts by TGM from the real texts, the latest detection models leverage learning algorithms with a variety of features to improve the detection performance against generated texts. However, the features included in these models are usually not relevant to the deep semantic correlation and fail to achieve accurate detection of the generated texts. To handle this problem, we propose a deep interpretable model to achieve accurate detection of the generated texts. In detail, a feature extraction method based on Layer-wise Relevance Propagation is proposed to improve the feature mining performance for detection of generated texts. Experiments based on real-world datasets demonstrate that the proposed model achieves more accurate generated text detection and significantly outperforms state-of-the-art detection schemes.

关键词： Analytical models Correlation Semantics Big data Feature extraction data models Fake news

来源：评论

学校读者我要写书评

暂无评论

Multi-view Event Recognition in Crowd Scenes using Convolutional Neural Networks 15

Multi-view Event Recognition in Crowd Scenes using Convoluti...

引用

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

作者： Garg, Avni Vanaja, T. Dev, Sudhanshu Kumar, R. Thanga Ingle, Yashwant S. Balaji, N.V. Chitkara University Institute of Engineering and Technology Chitkara University Punjab Rajpura140417 India Prince Shri Venkateshwara Padmavathy Engineering College Department of Computer Science and Engineering Chennai127 India Chitkara University Chitkara Centre for Research and Development Himachal Pradesh 174103 India School of Mangement - Ug Karnataka Bangalore India Vishwakarma Institute of Information Technology Department of Artificial Intelligence & Data Science Pune India Karpagam Academy of Higher Education Department of Computer Science Coimbatore641021 India

ISBN: (纸本)9798350370249

Convolutional Neural Networks (CNNs) have been gaining big interest in packages related to video reputation. This abstract focuses on their application to the recognition events in crowd scenes. Multi-view event popularity is proposed, with the CNN receiving a concatenated two-flow input contemporary pictures and optical waft-primarily based 3-D data from a couple of digi cam views. The CNN parameters are trained and tuned with the use of a multi-view low-stage feature aggregation level. The experiments performed on benchmark datasets demonstrate that the proposed approach gives advanced performance over current algorithms. The improved overall performance is attributed to the blended impact of state-of-the-art multi-view information and stepped forward CNN schooling. Moreover, it is verified that the low-level feature aggregation considerably improves the event's reputation and overall performance. © 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

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：