检索结果-内蒙古大学图书馆

The Conditional Cauchy-Schwarz Divergence With Applications to Time-Series data and Sequential Decision Making

学校读者我要写书评

暂无评论

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年第7期47卷 5901-5917页

作者： Shujian Yu Hongming Li Sigurd Løkse Robert Jenssen José C. Príncipe Machine Learning Group UiT - The Arctic University of Norway Tromsø Norway Quantitative Data Analytics Group Vrije Universiteit Amsterdam Amsterdam The Netherlands Department of Electrical and Computer Engineering University of Florida Gainesville FL USA Drones and Autonomous Systems Group NORCE Norwegian Research Centre Tromsø Norway Pioneer AI Centre Copenhagen University København Denmark Norwegian Computing Center Oslo Norway

The Cauchy-Schwarz (CS) divergence was developed by Príncipe et al. in 2000. In this paper, we extend the classic CS divergence to quantify the closeness between two conditional distributions and show that the developed conditional CS divergence can be elegantly estimated by a kernel density estimator from given samples. We illustrate the advantages (e.g., rigorous faithfulness guarantee, lower computational complexity, higher statistical power, and much more flexibility in a wide range of applications) of our conditional CS divergence over previous proposals, such as the conditional Kullback-Leibler divergence and the conditional maximum mean discrepancy. We also demonstrate the compelling performance of conditional CS divergence in two machine learning tasks related to time series data and sequential inference, namely time series clustering and uncertainty-guided exploration for sequential decision making.

关键词： Covariance matrices Kernel Machine learning Time series analysis Vectors Symmetric matrices Q measurement Matrix decomposition Estimation Decision making

Estimating Smart Grid Stability with Hybrid RNN+LSTM Deep Learning Approach

学校读者我要写书评

暂无评论

Estimating Smart Grid Stability with Hybrid RNN+LSTM Deep Le...

International Smart Grid Workshop and Certificate Program (ISGWCP)

作者： Saadin Oyucu Şeref Sağıroğlu Ahmet Aksöz Emre Biçer Department of Computer Engineering Faculty of Engineering Adıyaman Türkiye Gazi University Artificial Intelligence and Big Data Analytics Security R&D Center Ankara Türkiye Mobilers Team Sivas Cumhuriyet University Sivas Türkiye Faculty of Engineering and Natural Sciences Battery Research Laboratory Sivas University of Science and Technology Sivas Türkiye

ISBN: (数字)9798350361612

ISBN: (纸本)9798350361629

Smart grids are faced with a range of challenges, such as the development of communication infrastructure, cybersecurity threats, data privacy, and the protection of user information, due to their complex structure. Another key challenge faced by smart grids is the stability issues arising from variable energy sources and consumption patterns. In these complex grid systems where energy demand and supply need to be balanced instantly, stability predictions play a significant role in foreseeing potential disruptions and optimizing energy flow. Therefore, within the scope of this study, a hybrid structure utilizing Recurrent Neural Networks (RNN) and Long Short-Term Memory (LSTM) networks is employed for stability classification to predict grid stability. This hybrid model combines the ability of RNN to recognize relationships between consecutive data points with LSTM’s capability to preserve long-term dependencies. The results obtained indicate that the model exhibited stable performance with accuracy rates of 98.06% and 98.02% at 50 and 100 epochs, respectively. The findings of this study contribute valuable insights to research on the management and stability of smart grids, enabling energy systems to be operated more reliably and efficiently.

关键词： Deep learning data privacy Recurrent neural networks Power system stability Stability analysis data models Smart grids

An Improved Unbalanced data Classification Method Based on Hybrid Sampling Approach 4

学校读者我要写书评

暂无评论

An Improved Unbalanced Data Classification Method Based on H...

2021 IEEE 4th International Conference on big data and Artificial Intelligence, BDai 2021

作者： Xu, Biru Wang, Wenjia Yang, Rui Han, Qi University of Liverpool Department of Computer Science Liverpool United Kingdom School of Advanced Technology Research Institute of Big Data Analytics Xi an Jiaotong-Liverpool University Suzhou China College of Electrical Engineering and Automation Shandong University of Science and Technology Qingdao China

ISBN: (纸本)9781665412704

The problem of data imbalance has received far- reaching concerns since they could affect the accuracy of classification problem in the area of machine learning. As the minority class instances can be ignored by traditional classifiers, it is necessary to improve the recognition rate of minority instances. Therefore, the paper proposes a new hybrid sampling method to solve the data imbalance problem by enlarging the proportion of minority instances. For the oversampling part, a variant of SMOTE is provided combining methods of LR-SMOTE and CCR (Combined Cleaning and Resampling Algorithm);for the under-sampling part, the Tomek-link method is utilized to complete the task. After the pre-processing stage, the data set is classified by Random Forest (RF). Experimental results show that the novel algorithm effectively enhances the performance of RF on the data set with a higher accuracy. © 2021 IEEE.

关键词： data mining

A Hybrid Neural Network with Smart Skip Connections for High-Precision, Low-Latency EMG-Based Hand Gesture Recognition

学校读者我要写书评

暂无评论

A Hybrid Neural Network with Smart Skip Connections for High...

IEEE International Conference on Multi Topic

作者： Hafsa Wazir Jawad Ahmad Muazzam A. Khan Sana Ullah Jan Fadia Ali Khan Muhammad Shahbaz Khan ICESCO Chair for Big Data Analytics and Edge Computing Quaid-i-Azam University Islamabad Pakistan Cybersecurity Center Prince Mohammad Bin Fahd University Alkhobar Saudi Arabia School of Computing Engineering and the Built Environment Edinburgh Napier University Edinburgh UK Department of Computer Science HITEC University Taxila Pakistan

ISBN: (数字)9798331507213

ISBN: (纸本)9798331507220

Electromyography (EMG) is extensively used in key biomedical areas, such as prosthetics, and assistive and interactive technologies. EMG signals measure the electrical activity of muscles during different motions. EMG signals play a key role in gesture recognition studies, such as hand gesture recognition. This paper presents a new hybrid neural network named ConSGruNet for precise and efficient hand gesture recognition. The proposed model comprises convolutional neural networks with smart skip connections in conjunction with a Gated Recurrent Unit (GRU). The proposed model is trained on the complete Ninapro DB1 dataset. The proposed model boasts an accuracy of 99.7% in classifying 53 classes in just 25 milliseconds. In addition to being fast, the proposed model is lightweight with just 3,946 KB in size. The fast inference time and lightweight architecture of the proposed model makes it suitable for resource constrained IoT devices. Moreover, the proposed model has also been evaluated for the reliability parameters, i.e., Cohen’s kappa coefficient, Matthew’s correlation coefficient, and confidence intervals. The close to ideal results of these parameters validate the models performance on unseen data.

关键词： Hands Performance evaluation Correlation coefficient Accuracy Neural networks Gesture recognition Electromyography data models Internet of Things Reliability

Personalized Food Recommendations: A Machine Learning Model for Enhanced Dining Choices

学校读者我要写书评

暂无评论

Personalized Food Recommendations: A Machine Learning Model ...

IEEE International Conference on Control System, Computing and engineering (ICCSCE)

作者： Mohammed Al-Hubaishi Musab A. M. Ali Nooritawati Md Tahir Department of Computer Engineering Halic University Istanbul Turkey College of Computing and Information Sciences University of Technology and Applied Sciences Sultanate of Oman Department of Computer Engineering and Information Security International Information Technology University Kazakhstan Institute for Big Data Analytics and Artificial Intelligence (IBDAAI) Universiti Teknologi MARA Malaysia College of Engineering Universiti Teknologi MARA Malaysia Applied College Princess Nourah Bint Abdul Rahman University (PNU) Riyadh Saudi Arabia

ISBN: (数字)9798350364507

ISBN: (纸本)9798350364514

This research introduces an innovative approach to enhance dining experiences through a personalized meal suggestion system. Our method leverages decision tree-based machine learning to predict user preferences by analyzing various attributes. The aim is to assist individuals in making optimal food choices at restaurants, thereby enriching their culinary journey. QR codes are integrated into an Android restaurant application, facilitating convenient access to the food menu. User inquiries are efficiently managed through a MySQL database accessible across multiple platforms, including a Java GUI, an Android app, and a dedicated website catering to both user and administrator roles.

关键词： Java Codes databases Machine learning QR codes Control systems Decision trees Graphical user interfaces

Structure-Aware DropEdge Towards Deep Graph Convolutional Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Han, Jiaqi Huang, Wenbing Rong, Yu Xu, Tingyang Sun, Fuchun Huang, Junzhou Department of Computer Science and Technology Tsinghua University Beijing China Gaoling School of Artificial Intelligence Renmin University of China Beijing Key Laboratory of Big Data Management and Analysis Methods Beijing China Tencent AI Lab Shenzhen China Department of Computer Science and Engineering The University of Texas Arlington United States

It has been discovered that Graph Convolutional Networks (GCNs) encounter a remarkable drop in performance when multiple layers are piled up. The main factor that accounts for why deep GCNs fail lies in over-smoothing, which isolates the network output from the input with the increase of network depth, weakening expressivity and trainability. In this paper, we start by investigating refined measures upon DropEdge—an existing simple yet effective technique to relieve over-smoothing. We term our method as DropEdge++ for its two structure-aware samplers in contrast to DropEdge: layer-dependent sampler and feature-dependent sampler. Regarding the layer-dependent sampler, we interestingly find that increasingly sampling edges from the bottom layer yields superior performance than the decreasing counterpart as well as DropEdge. We theoretically reveal this phenomenon with Mean-Edge-Number (MEN), a metric closely related to over-smoothing. For the feature-dependent sampler, we associate the edge sampling probability with the feature similarity of node pairs, and prove that it further correlates the convergence subspace of the output layer with the input features. Extensive experiments on several node classification benchmarks, including both full- and semi- supervised tasks, illustrate the efficacy of DropEdge++ and its compatibility with a variety of backbones by achieving generally better performance over DropEdge and the no-drop version. Copyright © 2023, The Authors. All rights reserved.

关键词： Drops

MultiNet with Transformers: A Model for Cancer Diagnosis Using Images

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Barzekar, Hosein Patel, Yash Tong, Ling Yu, Zeyun Big Data Analytics and Visualization Laboratory Department of Computer Science University of Wisconsin-Milwaukee MilwaukeeWI53211 United States Department of Health Informatics and Administration University of Wisconsin-Milwaukee MilwaukeeWI53211 United States Department of Biomedical Engineering University of Wisconsin-Milwaukee MilwaukeeWI53211 United States

Cancer is a leading cause of death in many countries. An early diagnosis of cancer based on biomedical imaging ensures effective treatment and a better prognosis. However, biomedical imaging presents challenges to both clinical institutions and researchers. Physiological anomalies are often characterized by slight abnormalities in individual cells or tissues, making them difficult to detect visually. Traditionally, anomalies are diagnosed by radiologists and pathologists with extensive training. This procedure, however, demands the participation of professionals and incurs a substantial cost. The cost makes large-scale biological image classification impractical. In this study, we provide unique deep neural network designs for multiclass classification of medical images, in particular cancer images. We incorporated transformers into a multiclass framework to take advantage of data-gathering capability and perform more accurate classifications. We evaluated models on publicly accessible datasets using various measures to ensure the reliability of the models. Extensive assessment metrics suggest this method can be used for a multitude of classification tasks. © 2023, CC BY.

关键词： computer aided diagnosis

USED: Universal Speaker Extraction and Diarization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ao, Junyi Yıldırım, Mehmet Sinan Tao, Ruijie Ge, Meng Wang, Shuai Qian, Yanmin Li, Haizhou Shenzhen Research Institute of Big Data School of Data Science The Chinese University of Hong Kong Shenzhen518172 China Department of Electrical and Computer Engineering National University of Singapore Singapore119077 Singapore Saw Swee Hock School of Public Health National University of Singapore Singapore117549 Singapore Shenzhen Research Institute of Big Data Shenzhen518172 China Auditory Cognition and Computational Acoustics Lab Department of Computer Science and Engineering China MoE Key Laboratory of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai200240 China

Speaker extraction and diarization are two enabling techniques for real-world speech applications. Speaker extraction aims to extract a target speaker’s voice from a speech mixture, while speaker diarization demarcates speech segments by speaker, annotating ‘who spoke when’. Previous studies have typically treated the two tasks independently. In practical applications, it is more meaningful to have knowledge about ‘who spoke what and when’, which is captured by the two tasks. The two tasks share a similar objective of disentangling speakers. Speaker extraction operates in the frequency domain, whereas diarization is in the temporal domain. It is logical to believe that speaker activities obtained from speaker diarization can benefit speaker extraction, while the extracted speech offers more accurate speaker activity detection than the speech mixture. In this paper, we propose a unified model called Universal Speaker Extraction and Diarization (USED) to address output inconsistency and scenario mismatch issues. It is designed to manage speech mixtures with varying overlap ratios and variable number of speakers. We show that the USED model significantly outperforms the competitive baselines for speaker extraction and diarization tasks on LibriMix and SparseLibriMix datasets. We further validate the diarization performance on CALLHOME, a dataset based on real recordings, and experimental results indicate that our model surpasses recently proposed approaches. Copyright © 2023, The Authors. All rights reserved.

关键词： Frequency domain analysis

DISENTANGLING THE PROSODY AND SEMANTIC INFORMATION WITH PRE-TRaiNED MODEL FOR IN-CONTEXT LEARNING BASED ZERO-SHOT VOICE CONVERSION

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chen, Zhengyang Wang, Shuai Zhang, Mingyang Liu, Xuechen Yamagishi, Junichi Qian, Yanmin Auditory Cognition and Computational Acoustics Lab MoE Key Lab of Artificial Intelligence AI Institute Department of Computer Science and Engineering Shanghai Jiao Tong University Shanghai China Shenzhen Research Institute of Big Data Shenzhen China Chinese University of Hong Kong Shenzhen China National Institute of Informatics Tokyo Japan

Voice conversion (VC) aims to modify the speaker's timbre while retaining speech content. Previous approaches have tokenized the outputs from self-supervised into semantic tokens, facilitating disentanglement of speech content information. Recently, in-context learning (ICL) has emerged in text-to-speech (TTS) systems for effectively modeling specific characteristics such as timbre through context conditioning. This paper proposes an ICL capability enhanced VC system (ICL-VC) employing a mask and reconstruction training strategy based on flow-matching generative models. Augmented with semantic tokens, our experiments on the LibriTTS dataset demonstrate that ICL-VC improves speaker similarity. Additionally, we find that k-means is a versatile tokenization method applicable to various pre-trained models. However, the ICL-VC system faces challenges in preserving the prosody of the source speech. To mitigate this issue, we propose incorporating prosody embeddings extracted from a pre-trained emotion recognition model into our system. Integration of prosody embeddings notably enhances the system's capability to preserve source speech prosody, as validated on the Emotional Speech database. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics