检索结果-内蒙古大学图书馆

Identifying the Complete Correlation Structure in Large-Scale High-Dimensional data Sets with Local False Discovery Rates

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Gölz, Martin Hasija, Tanuj Muma, Michael Zoubir, Abdelhak M. The Signal Processing Group TU Darmstadt Germany The Signal and Systems Group Paderborn University Germany The Robust Data Science Group TU Darmstadt Germany

The identification of the dependent components in multiple data sets is a fundamental problem in many practical applications. The challenge in these applications is that often the data sets are high-dimensional with few observations or available samples and contain latent components with unknown probability distributions. A novel mathematical formulation of this problem is proposed, which enables the inference of the underlying correlation structure with strict false positive control. In particular, the false discovery rate is controlled at a pre-defined threshold on two levels simultaneously. The deployed test statistics originate in the sample coherence matrix. The required probability models are learned from the data using the bootstrap. Local false discovery rates are used to solve the multiple hypothesis testing problem. Compared to the existing techniques in the literature, the developed technique does not assume an a priori correlation structure and work well when the number of data sets is large while the number of observations is small. The simulation results underline that it can handle the presence of distributional uncertainties, heavy-tailed noise, and outliers. Copyright © 2023, The Authors. All rights reserved.

关键词： Probability distributions

A Unified Framework Integrating Knowledge and data for Collaborative Root Cause Identification

学校读者我要写书评

暂无评论

A Unified Framework Integrating Knowledge and Data for Colla...

Sensing, Measurement & data Analytics in the era of Artificial Intelligence (ICSMD), International Conference on

作者： Jiefei Yu Zicheng Cao Siyi He Zuyi Gu Yingcheng Xu Kai Zhong Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education Institutes of Physical Science and Information Technology Anhui University Anhui China Data Science and Big Data Technology School of Big Data and Statistics Anhui University Anhui China

ISBN: (数字)9798331529192

ISBN: (纸本)9798331529208

Capturing the root cause and propagation path of the fault is critical to ensuring the safety and efficiency of industrial processes, especially those that inadequately utilize process knowledge and data. To address this issue, a unified framework integrating knowledge and data for collaborative root cause identification is proposed. First, the knowledge causal graph (KCG) is constructed using expert knowledge and industrial flow charts, providing a preliminary reference for subsequent causality analysis. Next, by replacing the traditional vector autoregression (VAR) model in Granger Causality (GC) with the gated recurrent unit (GRU), a more reliable causal relationship between variables is obtained. Additionally, a causality fusion propagation path identification method (CF-PPI) is designed to identify the root cause and propagation path of the fault, so that the obtained fault propagation path has less redundancy and higher accuracy. Finally, the method is validated using data from the ASHRAE RP-1043 centrifugal chiller.

关键词： Fault diagnosis Knowledge engineering Reactive power Analytical models Delay effects Redundancy Collaboration Cause effect analysis Vectors Safety

MetaAug: Meta-data Augmentation for Post-Training Quantization

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Pham, Cuong Dung, Hoang Anh Nguyen, Cuong C. Le, Trung Phung, Dinh Carneiro, Gustavo Do, Thanh-Toan Department of Data Science and AI Monash University Australia Centre for Vision Speech and Signal Processing University of Surrey United Kingdom VinAI Viet Nam

Post-Training Quantization (PTQ) has received significant attention because it requires only a small set of calibration data to quantize a full-precision model, which is more practical in real-world applications in which full access to a large training set is not available. However, it often leads to overfitting on the small calibration dataset. Several methods have been proposed to address this issue, yet they still rely on only the calibration set for the quantization and they do not validate the quantized model due to the lack of a validation set. In this work, we propose a novel meta-learning based approach to enhance the performance of post-training quantization. Specifically, to mitigate the overfitting problem, instead of only training the quantized model using the original calibration set without any validation during the learning process as in previous PTQ works, in our approach, we both train and validate the quantized model using two different sets of images. In particular, we propose a meta-learning based approach to jointly optimize a transformation network and a quantized model through bi-level optimization. The transformation network modifies the original calibration data and the modified data will be used as the training set to learn the quantized model with the objective that the quantized model achieves a good performance on the original calibration data. Extensive experiments on the widely used ImageNet dataset with different neural network architectures demonstrate that our approach outperforms the state-of-the-art PTQ methods. Code is available at this https URL. © 2024, CC BY.

关键词： Contrastive Learning

A Joint Learning Sentiment Analysis Method Incorporating Emoji-Augmentation 8

学校读者我要写书评

暂无评论

A Joint Learning Sentiment Analysis Method Incorporating Emo...

8th IEEE International Conference on Cloud Computing and Intelligence Systems, CCIS 2022

作者： Chen, Jie Luo, Luping Ji, Bojing Zhao, Shu Zhang, Yanping Anhui University The Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology China Chizhou University School of Big Data and Artificial Intelligence China

ISBN: (纸本)9781665477352

Social media is the platform for most people to share their opinions, emojis are also widely used to express moods, emotions, and feelings on social media. There have been many researched on emojis and sentiment analysis. However, existing methods mainly face two limitations. First, since deep learning relies on large amounts of labeled data, the training samples of emoji are not enough to achieve the training effect. Second, they consider the sentiment of emojis and texts separately, not fully exploring the impact of emojis on the sentiment polarity of texts. In this paper, we propose a joint learning sentiment analysis method incorporating emoji-augmentation, and the method has two advantages compared with the existing work. First, We optimize the easy data augmentation method so that the newly generated sentences can also preserve the semantic information of emojis, which relieves the problem of insufficient training data with emojis. Second, it fuses emojis and text features to allow the model to better learn the mutual emotional semantics between text and emojis, jointly training emojis and words to obtain the sentence representations containing more semantic information of both emojis and text. Our experimental results show that the proposed method can significantly improve the performance compared with several baselines on two datasets. © 2022 IEEE.

关键词： Sentiment analysis

Heterogeneous-attributes enhancement deep framework for network embedding

学校读者我要写书评

暂无评论

Frontiers of Computer science 2021年第6期15卷 121-131页

作者： Lisheng QIAO Fan ZHANG Xiaohui HUANG Kai LI Enhong CHEN Anhui Province Key Laboratory of Big Data Analysis and Application University of Science and Technology of ChinaHefei 230022China National Key Laboratory of Blind Signal Processing Chengdu 610041China School of Computer Science and Technology University of Science and Technology of ChinaHefei 230022China

Network embedding,which targets at learning the vector representation of vertices,has become a crucial issue in network ***,considering the complex structures and heterogeneous attributes in real-world networks,existing methods may fail to handle the inconsistencies between the structure topology and attribute ***,more comprehensive techniques are urgently required to capture the highly non-linear network structure and solve the existing inconsistencies with retaining more *** that end,in this paper,we propose a heterogeneous-attributes enhancement deep framework(HEDF),which could better capture the non-linear structure and associated information in a deep learningway,and effectively combine the structure information of multi-views by the combining *** this line,the inconsistencies will be handled to some extent and more structure information will be preserved through a semi-supervised *** extensive validations on several real-world datasets show that our model could outperform the baselines,especially for the sparse and inconsistent situation with less training data.

关键词： network embedding heterogeneous-attributes deep framework inconsistent

Learning to Complement with Multiple Humans

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Zhang, Zheng Nguyen, Cuong Wells, Kevin Do, Thanh-Toan Carneiro, Gustavo Centre for Vision Speech and Signal Processing University of Surrey United Kingdom Department of Data Science and AI Monash University Australia

Real-world image classification tasks tend to be complex, where expert labellers are sometimes unsure about the classes present in the images, leading to the issue of learning with noisy labels (LNL). The ill-posedness of the LNL task requires the adoption of strong assumptions or the use of multiple noisy labels per training image, resulting in accurate models that work well in isolation but fail to optimise human-AI collaborative classification (HAI-CC). Unlike such LNL methods, HAI-CC aims to leverage the synergies between human expertise and AI capabilities but requires clean training labels, limiting its real-world applicability. This paper addresses this gap by introducing the innovative Learning to Complement with Multiple Humans (LECOMH) approach. LECOMH is designed to learn from noisy labels without depending on clean labels, simultaneously maximising collaborative accuracy while minimising the cost of human collaboration, measured by the number of human expert annotations required per image. Additionally, new benchmarks featuring multiple noisy labels for both training and testing are proposed to evaluate HAI-CC methods. Through quantitative comparisons on these benchmarks, LECOMH consistently outperforms competitive HAI-CC approaches, human labellers, multi-rater learning, and noisy-label learning methods across various datasets, offering a promising solution for addressing real-world image classification challenges. © 2023, CC BY-NC-ND.

关键词： Image classification

An External Denoising Framework for Magnetic Resonance Imaging: Leveraging Anatomical Similarities Across Subjects with Fast Searches 8

学校读者我要写书评

暂无评论

An External Denoising Framework for Magnetic Resonance Imagi...

8th International Conference on signal and Image processing, ICSIP 2023

作者： Mei, Lifeng Liu, Sixing Tang, Chenhui Cai, Jiaqia Wang, Jingli Liu, Yilong Wu, Ed X. Lyu, Mengye Shenzhen Technology University College of Health Science and Environmental Engineering Shenzhen China Shenzhen Technology University College of Big Data and Internet Shenzhen China Guangzhou China The University of Hong Kong Laboratory of Biomedical Imaging and Signal Processing Department of Electrical and Electronic Engineering Hong Kong Hong Kong

ISBN: (纸本)9798350397932

External denoising, also known as reference-based denoising, utilizes information from clean reference images, yielding more robust results than internal denoising, especially in situations with high noise levels. The effectiveness of reference-based denoising relies on the similarity between reference and noisy images. In this paper, we introduce a novel external denoising framework for magnetic resonance imaging (MRI) that takes advantage of the inherent anatomical structure similarities across subjects. Our framework employs noise-resistant neural networks to extract deep features, which facilitate searches for high-quality reference images from a large external database. This versatile framework is compatible with all reference-based denoising algorithms, obviating the need for acquiring additional reference images from the subject under examination. © 2023 IEEE.

关键词： Magnetic resonance imaging

Coverage-Constrained Human-AI Cooperation with Multiple Experts

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Zheng Nguyen, Cuong Wells, Kevin Do, Toan Rosewarne, David Carneiro, Gustavo Centre for Vision Speech and Signal Processing University of Surrey United Kingdom Department of Data Science and AI Monash University Australia Royal Wolverhampton Hospitals NHS Trust United Kingdom

Human-AI cooperative classification (HAI-CC) approaches aim to develop hybrid intelligent systems that enhance decision-making in various high-stakes real-world scenarios by leveraging both human expertise and AI capabilities. Current HAI-CC methods primarily focus on learning-to-defer (L2D), where decisions are deferred to human experts, and learning-to-complement (L2C), where AI and human experts make predictions cooperatively. However, a notable research gap remains in effectively exploring both L2D and L2C under diverse expert knowledge to improve decision-making, particularly when constrained by the cooperation cost required to achieve a target probability for AI-only selection (i.e., coverage). In this paper, we address this research gap by proposing the Coverage-constrained Learning to Defer and Complement with Specific Experts (CL2DC) method. CL2DC makes final decisions through either AI prediction alone or by deferring to or complementing a specific expert, depending on the input data. Furthermore, we propose a coverage-constrained optimisation to control the cooperation cost, ensuring it approximates a target probability for AI-only selection. This approach enables an effective assessment of system performance within a specified budget. Also, CL2DC is designed to address scenarios where training sets contain multiple noisy-label annotations without any clean-label references. Comprehensive evaluations on both synthetic and real-world datasets demonstrate that CL2DC achieves superior performance compared to state-of-the-art HAI-CC methods. © 2024, CC BY.

关键词： Budget control

Pass: Peer-Agreement Based Sample Selection for Training with Instance-Dependent Noisy Labels

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Garg, Arpit Nguyen, Cuong Felix, Rafael Do, Thanh-Toan Carneiro, Gustavo Australian Institute for Machine Learning University of Adelaide Australia Centre for Vision Speech and Signal Processing University of Surrey United Kingdom Department of Data Science and AI Monash University Australia

The prevalence of noisy-label samples poses a significant challenge in deep learning, inducing overfitting effects. This has, therefore, motivated the emergence of learning with noisy-label (LNL) techniques that focus on separating noisy- and clean-label samples to apply different learning strategies to each group of samples. Current methodologies often rely on the small-loss hypothesis or feature-based selection to separate noisy- and clean-label samples, yet our empirical observations reveal their limitations, especially for labels with instance dependent noise (IDN). An important characteristic of IDN is the difficulty to distinguish the clean-label samples that lie near the decision boundary (i.e., the hard samples) from the noisy-label samples. We, therefore, propose a new noisy-label detection method, termed Peer-Agreement based Sample Selection (PASS), to address this problem. Utilising a trio of classifiers, PASS employs consensus-driven peer-based agreement of two models to select the samples to train the remaining model. PASS is easily integrated into existing LNL models, enabling the improvement of the detection accuracy of noisy- and clean-label samples, which increases the classification accuracy across various LNL benchmarks. © 2024, The Authors. All rights reserved.

关键词： Deep learning