检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Gill, Sukhpal Singh Wu, Huaming Patros, Panos Ottaviani, Carlo Arora, Priyansh Pujol, Victor Casamayor Haunschild, David Parlikad, Ajith Kumar Cetinkaya, Oktay Lutfiyya, Hanan Stankovski, Vlado Li, Ruidong Ding, Yuemin Qadir, Junaid Abraham, Ajith Ghosh, Soumya K. Song, Houbing Herbert Sakellariou, Rizos Rana, Omer Rodrigues, Joel J.P.C. Kanhere, Salil S. Dustdar, Schahram Uhlig, Steve Ramamohanarao, Kotagiri Buyya, Rajkumar School of Electronic Engineering and Computer Science Queen Mary University of London London United Kingdom Center for Applied Mathematics Tianjin University Tianjin China Raygun Performance Monitoring Wellington New Zealand Department of Computer Science York Centre for Quantum Technologies University of York York United Kingdom Microsoft Schiphol Netherlands Distributed Systems Group Vienna University of Technology Vienna Austria Detecon International GmbH Munich Germany Institute for Manufacturing Department of Engineering University of Cambridge Cambridge United Kingdom Department of Engineering Science University of Oxford Oxford United Kingdom Department of Computer Science University of Western Ontario London Canada Faculty of Computer and Information Science University of Ljubljana Ljubljana Slovenia Institute of Science and Engineering Kanazawa University Japan Tecnun School of Engineering University of Navarra Spain Department of Computer Science and Engineering College of Engineering Qatar University Doha Qatar Bennett University Greater Noida India Machine Intelligence Research Labs AuburnWA United States Department of Computer Science and Engineering Indian Institute of Technology Kharagpur India Baltimore United States Department of Computer Science University of Manchester Oxford Road Manchester United Kingdom School of Computer Science and Informatics Cardiff University Cardiff United Kingdom Department of Computer Science College of Computer and Information Sciences King Saud University Riyadh Saudi Arabia COPELABS Lusófona University Lisbon Portugal Sydney Australia The University of Melbourne Victoria Australia Laboratory School of Computing and Information Systems The University of Melbourne Australia

Over the past six decades, the computing systems field has experienced significant transformations, profoundly impacting society with transformational developments, such as the Internet and the commodification of computing. Underpinned by technological advancements, computer systems, far from being static, have been continuously evolving and adapting to cover multifaceted societal niches. This has led to new paradigms such as cloud, fog, edge computing, and the Internet of Things (IoT), which offer fresh economic and creative opportunities. Nevertheless, this rapid change poses complex research challenges, especially in maximizing potential and enhancing functionality. As such, to maintain an economical level of performance that meets ever-tighter requirements, one must understand the drivers of new model emergence and expansion, and how contemporary challenges differ from past ones. To that end, this article investigates and assesses the factors influencing the evolution of computing systems, covering established systems and architectures as well as newer developments, such as serverless computing, quantum computing, and on-device AI on edge devices. Trends emerge when one traces technological trajectory, which includes the rapid obsolescence of frameworks due to business and technical constraints, a move towards specialized systems and models, and varying approaches to centralized and decentralized control. This comprehensive review of modern computing systems looks ahead to the future of research in the field, highlighting key challenges and emerging trends, and underscoring their importance in cost-effectively driving technological progress. © 2024, CC BY.

关键词： Obsolescence

来源：评论

学校读者我要写书评

暂无评论

Your Transformer May Not be as Powerful as You Expect

arXiv

引用

arXiv 2022年

作者： Luo, Shengjie Li, Shanda Zheng, Shuxin Liu, Tie-Yan Wang, Liwei He, Di Key Laboratory of Machine Perception MOE School of Intelligence Science and Technology Peking University China Machine Learning Department School of Computer Science Carnegie Mellon University United States Microsoft Research United States Center for Data Science Peking University China Zhejiang Lab China

Relative Positional Encoding (RPE), which encodes the relative distance between any pair of tokens, is one of the most successful modifications to the original Transformer. As far as we know, theoretical understanding of the RPE-based Transformers is largely unexplored. In this work, we mathematically analyze the power of RPE-based Transformers regarding whether the model is capable of approximating any continuous sequence-to-sequence functions. One may naturally assume the answer is in the affirmative-RPE-based Transformers are universal function approximators. However, we present a negative result by showing there exist continuous sequence-to-sequence functions that RPE-based Transformers cannot approximate no matter how deep and wide the neural network is. One key reason lies in that most RPEs are placed in the softmax attention that always generates a right stochastic matrix. This restricts the network from capturing positional information in the RPEs and limits its capacity. To overcome the problem and make the model more powerful, we first present sufficient conditions for RPE-based Transformers to achieve universal function approximation. With the theoretical guidance, we develop a novel attention module, called Universal RPE-based (URPE) Attention, which satisfies the conditions. Therefore, the corresponding URPE-based Transformers become universal function approximators. Extensive experiments covering typical architectures and tasks demonstrate that our model is parameter-efficient and can achieve superior performance to strong baselines in a wide range of applications. The code will be made publicly available at https://***/lsj2408/URPE. Copyright © 2022, The Authors. All rights reserved.

关键词： Stochastic systems

来源：评论

学校读者我要写书评

暂无评论

SEFA: A Shared Encoder and Feature Adaptation Framework for Breast MRI Segmentation

SEFA: A Shared Encoder and Feature Adaptation Framework for ...

引用

Complex Systems and Intelligent science (CSIS-IAC), International Annual Conference on

作者： Hongwei Xue Ke Zhong Xiaofen Wu Yan Gao Yujiao Wu Lei Zhang Guangwu Qian Peijun Wang Department of Radiology Tongji Hospital School of Medicine Tongji University Shanghai China Machine Intelligence Laboratory College of Computer Science Sichuan University Chengdu China The Commonwealth Scientific and Industrial Research Organisation Hobart Australia Sichuan University-Pittsburgh Institute Sichuan University Chengdu China

Significant progress has been made in developing accurate automatic segmentation systems for various biomedical applications using convolutional networks (CNNs). However, these systems often lose effectiveness when they encounter a domain shift caused by variations in imaging protocols. Traditional supervised transfer learning approaches are not ideal because manually annotating new data for every testing domain is impractical, while unsupervised domain adaptation is a challenging subject in the field of biomedical image analysis. In this paper, we introduce a new framework for unsupervised domain adaptation called Shared Encoder and Feature Adaptation (SEFA). This framework focuses on cross-sequences adaptation and is more resilient to disparities in input data and does not require any annotations on the test domain. Specifically, our segmenter model is based on a compact fully convolutional network designed for breast mask prediction. First, the shared encoder is designed to minimize the distribution disparity between the source and target domains. Then, a domain discriminator is set up to discriminate the feature space of both domains. Our model is optimized by using unpaired TSE/DCE sequences in an unsupervised manner. This eliminates the need for labeling additional medical datasets. Using our unsupervised approach, we achieve segmentation accuracies that are comparable to those obtained through supervised training. These accuracies are measured by metrics such as Case/Global Dice Score and average symmetric surface distance (ASD), which are close to the best possible performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

I Can’t Believe It’s Not Better: In-air Movement For Alzheimer Handwriting Synthetic Generation

arXiv

引用

arXiv 2023年

作者： Bensalah, Asma Parziale, Antonio De Gregorio, Giuseppe Marcelli, Angelo Fornés, Alicia Lladós, Josep Computer Vision Center Universitat Autònoma de Barcelona Spain Computer Science Department Universitat Autònoma de Barcelona Spain DIEM University of Salerno Via Giovanni Paolo II 132 SA Fisciano84084 Italy AI3S Unit CINI National Laboratory of Artificial Intelligence and Intelligent Systems University of Salerno SA Fisciano Italy

During recent years, there here has been a boom in terms of deep learning use for handwriting analysis and recognition. One main application for handwriting analysis is early detection and diagnosis in the health field. Unfortunately, most real case problems still suffer a scarcity of data, which makes difficult the use of deep learning-based models. To alleviate this problem, some works resort to synthetic data generation. Lately, more works are directed towards guided data synthetic generation, a generation that uses the domain and data knowledge to generate realistic data that can be useful to train deep learning models. In this work, we combine the domain knowledge about the Alzheimer’s disease for handwriting and use it for a more guided data generation. Concretely, we have explored the use of in-air movements for synthetic data generation. © 2023, CC BY.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

An Ensemble Approach to Multi-Class Classification of Vocal Disorders: Laryngocele and Vox Senilis

An Ensemble Approach to Multi-Class Classification of Vocal ...

引用

Intelligent Signal Processing and Effective Communication Technologies (INSPECT), IEEE International Conference on

作者： Puneet Bawa Virender Kadyan Archana Mantri Monika Sethi Centre of Excellence for Speech and Multimodal Laboratory Chitkara University Institute of Engineering & Technology Chitkara University Punjab India Machine Intelligence Research Centre School of Computer Science UPES Dehradun Uttarakhand India Department of Electronicsand Communication Engineering Anurag University Hyderabad India Chitkara University Institute of Engineering & Technology Chitkara University Punjab India

ISBN: (数字)9798350379525

ISBN: (纸本)9798350379532

The classification of audio signals has been a significant challenge in machine learning, especially with regard to the early identification of voice disorders. However, traditional techniques based on raw audio feature extraction have been shown to be not as effective due to the complex and non-stationary capacities of individuals experiencing speech difficulties. Therefore, the audio information related to two vocal disorders including Laryn-gocele and Vox Senilis have been transformed into an informative representation, Mel-Spectrogram. This helps in capturing both temporal as well as spectral characteristics in a manner consistent with typical auditory perception. In this study, the standalone Convolutional Neural Network (CNN) and various ensemble methods in conjection with machine learning techniques (CNN-ML) models have been explored for classification of audio signals based on Mel-Spectrogram. The CNN-Support Vector machine (SVM) based ensemble method has excelled with accuracy of 96.23% at the extraction of hierarchical features corresponding to the Mel-Spectrograms while the SVM component provides robust classification capabilities. Overall, a relative improvement of 41.18% has been observed for CNN-SVM using Adamax optimizer when compared to the performance of ensemble techniques using Adam optimizer.

关键词： Support vector machines Performance evaluation Accuracy Feature extraction Vectors Convolutional neural networks Ensemble learning Tuning Optimization Diseases

来源：评论

学校读者我要写书评

暂无评论

Improved Differential Evolution based Feature Selection through Quantum, Chaos, and Lasso

arXiv

引用

arXiv 2024年

作者： Vivek, Yelleti Vadlamani, Sri Krishna Ravi, Vadlamani Krishna, P. Radha Centre for Artificial Intelligence and Machine Learning Institute for Development and Research in Banking Technology Castle Hills Road #1 Masab Tank Telangana Hyderabad500076 India Research Laboratory of Electronics Massachusetts Institute of Technology CambridgeMA United States Department of Computer Science and Engineering National Institute of Technology Warangal Telangana Warangal506004 India

Modern deep learning continues to achieve outstanding performance on an astounding variety of high-dimensional tasks. In practice, this is obtained by fitting deep neural models to all the input data with minimal feature engineering, thus sacrificing interpretability in many cases. However, in applications such as medicine, where interpretability is crucial, feature subset selection becomes an important problem. Metaheuristics such as Binary Differential Evolution are a popular approach to feature selection, and the research literature continues to introduce novel ideas, drawn from quantum computing and chaos theory, for instance, to improve them. In this paper, we demonstrate that introducing chaos-generated variables, generated from considerations of the Lyapunov time, in place of random variables in quantum-inspired metaheuristics significantly improves their performance on high-dimensional medical classification tasks and outperforms other approaches. We show that this chaos-induced improvement is a general phenomenon by demonstrating it for multiple varieties of underlying quantum-inspired metaheuristics. Performance is further enhanced through Lasso-assisted feature pruning. At the implementation level, we vastly speed up our algorithms through a scalable island-based computing cluster parallelization *** Codes 68W50, 90C27 © 2024, CC BY-NC-ND.

关键词： Quantum chaos

来源：评论

学校读者我要写书评

暂无评论

Towards Disease-Aware Self-Supervised Dynamic Brain Network Learning For Mental Diagnosis

Towards Disease-Aware Self-Supervised Dynamic Brain Network ...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Zhiyong Jin Guangqi Wen Peng Cao Lingwen Liu Jinzhu Yang Xinrong Zhu Osmar R. Zaiane Fei Wang Computer Science and Engineering Northeastern University Shenyang China Key Laboratory of Intelligent Computing in Medical Image of Ministry of Education Northeastern University Shenyang China Department of Psychiatry Affiliated Nanjing Brain Hospital Early Intervention Unit Nanjing Medical University Nanjing China Alberta Machine Intelligence Institute University of Alberta Edmonton Alberta Canada

The dynamic brain network learning methods ignored the separation of redundant disease-irrelevant information, resulting in the model only achieving suboptimal diagnosis results. Meanwhile, the supervised learning scheme inevitably suffers from poor generalization due to the limited data. To address these problems, we propose a Self-supervised Dynamic Brain network Disentangled representation learning framework named SDBD, which incorporates 1) a dynamic topology-aware encoder for capturing diverse topological information, 2) a cross decoder for reconstructing the graph structure and 3) a spatio-temporal learning model based on the multi-head self-attention mechanism for classification. To disentangle the disease-related information from the dynamic brain networks, we design a temporal contrastive loss and a structure reconstruction loss. We evaluate our model on three real-world mental diseases: Autism Spectrum Disorder (ASD), Major Depressive Disorder (MDD), and Bipolar Disorder (BD). The results indicate significant improvements in our SDBD over the state-of-the-art methods owing to the disentangled disease-related information. Moreover, our method can identify the biomarkers associated with the diseases, which is consistent with the previous studies. To the best of our knowledge, our work is the first attempt to disentangle the disease-related information for the dynamic brain network analysis. The code is available at https://***/IntelliDAL/Graph/tree/main/SDBD.

关键词：

来源：评论

学校读者我要写书评

暂无评论

NavCoT: Boosting LLM-Based vision-and-Language Navigation via Learning Disentangled Reasoning

引用

IEEE Transactions on Pattern Analysis and machine intelligence 2025年第7期47卷 5945-5957页

作者： Bingqian Lin Yunshuang Nie Ziming Wei Jiaqi Chen Shikui Ma Jianhua Han Hang Xu Xiaojun Chang Xiaodan Liang Shenzhen Campus Sun Yat-sen University Shenzhen China Shanghai Jiao Tong University Shanghai China University of Hong Kong Pok Fu Lam Hong Kong Dataa Robotics Company Beijing China Huawei Noah's Ark Lab Shanghai China School of Information Science and Technology University of Science and Technology of China Hefei China Department of Computer Vision Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) Abu Dhabi UAE Peng Cheng Laboratory Shenzhen China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China

vision-and-Language Navigation (VLN), as a crucial research problem of Embodied AI, requires an embodied agent to navigate through complex 3D environments following natural language instructions. Recent research has highlighted the promising capacity of large language models (LLMs) in VLN by improving navigational reasoning accuracy and interpretability. However, their predominant use in an offline manner usually suffers from substantial domain gap between the VLN task and the LLM training corpus. This paper proposes a novel strategy called Nav igational C hain- o f- T hought (NavCoT), where we fulfill parameter-efficient in-domain training to enable self-guided navigational decision, leading to a significant mitigation of the domain gap in a cost-effective manner. Specifically, at each timestep, the LLM is prompted to forecast the navigational chain-of-thought by: 1) acting as a world model to imagine the next observation according to the instruction, 2) selecting the candidate observation that best aligns with the imagination, and 3) determining the action based on the reasoning from the prior steps. In this way, the action prediction can be effectively simplified benefiting from the disentangled reasoning. Through constructing formalized labels for training, the LLM can learn to generate desired and reasonable chain-of-thought outputs for improving the action decision. Experimental results across various training settings and popular VLN benchmarks (e.g., Room-to-Room (R2R), Room-across-Room (RxR), Room-for-Room (R4R)) show the significant superiority of NavCoT over the direct action prediction variants. Through simple parameter-efficient finetuning, our NavCoT outperforms a recent GPT4-based approach with $\sim$ 7% relative improvement on the R2R dataset. We believe that NavCoT will help unlock more task-adaptive and scalable LLM-based embodied agents, which are helpful for developing real-world robotics applications.

关键词： Navigation Cognition Training Scalability Glass Planning Large language models History Electronic mail Visualization

来源：评论

学校读者我要写书评

暂无评论

Clinical efficacy and safety of vortioxetine as an adjuvant drug for patients with bipolar depression

引用

Journal of Zhejiang University-science B(Biomedicine & Biotechnology) 2025年第1期26卷 26-38页

作者： Chunxiao DAI Yaoyang FU Xuanwei LI Meihua LIN Yinbo LI Xiao LI Keke HUANG Chengcheng ZHOU Jian XIE Qingwei ZHAO Shaohua HU Department of Psychiatry the First Affiliated HospitalZhejiang University School of MedicineHangzhou 310003China Department of Clinical Psychology Hangzhou First People’s HospitalHangzhou 310006China Department of Clinical Pharmacy the First Affiliated HospitalZhejiang University School of MedicineHangzhou 310003China Fuyang Third People’s Hospital Hangzhou 311402China Nanhu Brain-Computer Interface Institute Hangzhou 311100China The Zhejiang Key Laboratory of Precision Psychiatry Hangzhou 310003China School of Brain Science and Brain Medicine and MOE Frontier Science Center for Brain Science and Brain-Machine IntegrationZhejiang University School of MedicineHangzhou 310012China Brain Research Institute of Zhejiang University Hangzhou 310003China The State Key Lab of Brain-Machine Intelligence Zhejiang UniversityHangzhou 310003China Zhejiang Engineering Center for Mathematical Mental Health Hangzhou 310003China

Objective:Whether vortioxetine has a utility as an adjuvant drug in the treatment of bipolar depression remains *** study aimed to validate the efficacy and safety of vortioxetine in bipolar ***:Patients with bipolarⅡdepression were enrolled in this prospective,two-center,randomized,12-week pilot *** main indicator for assessing treatment effectiveness was a Montgomery-Asberg Depression Rating Scale(MADRS)of≥50%.All eligible patients initially received four weeks of lurasidone *** who responded well continued to receive this kind of ***,no-response patients were randomly assigned to either valproate or vortioxetine treatment for eight *** comprehensively comparing the results of MADRS over a period of 4-12 weeks,a systematic analysis was conducted to determine whether vortioxetine could be used as an adjuvant drug for treating bipolar ***:Thirty-seven patients responded to lurasidone monotherapy,and 60 patients were randomly assigned to the valproate or vortioxetine group for eight *** two weeks of combined valproate or vortioxetine treatment,the MADRS score in the vortioxetine group was significantly lower than that in the valproate *** was no difference in the MADRS scores between the two groups at 8 and 12 *** incidence of side effects did not significantly differ between the valproate and vortioxetine ***,three patients in the vortioxetine group appeared to switch to mania or ***:This study suggested that lurasidone combination with vortioxetine might have potential benefits to bipolar II depression in the early stage,while disease progression should be monitored closely for the risk of switching to mania.

关键词： BipolarⅡdepression Lurasidone Vortioxetine Combination

来源：评论

学校读者我要写书评

暂无评论

GRATIS: Deep Learning Graph Representation with Task-specific Topology and Multi-dimensional Edge Features

arXiv

引用

arXiv 2022年

作者： Song, Siyang Song, Yuxin Luo, Cheng Song, Zhiyuan Kuzucu, Selim Jia, Xi Guo, Zhijiang Xie, Weicheng Shen, Linlin Gunes, Hatice The Department of Computer Science and Technology University of Cambridge CambridgeCB3 0FT United Kingdom Baidu Inc Beijing100193 China Computer Vision Institute School of Computer Science & Software Engineering Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518060 China Department of Computer Engineering Middle East Technical University Ankara Turkey School of Computer Science University of Birmingham Birmingham United Kingdom

Graph is powerful for representing various types of real-world data. The topology (edges’ presence) and edges’ features of a graph decides the message passing mechanism among vertices within the graph. While most existing approaches only manually define a single-value edge to describe the connectivity or strength of association between a pair of vertices, task-specific and crucial relationship cues may be disregarded by such manually defined topology and single-value edge features. In this paper, we propose the first general graph representation learning framework (called GRATIS) which can generate a strong graph representation with a task-specific topology and task-specific multi-dimensional edge features from any arbitrary input (i.e., graph or non-graph data). To learn each edge’s presence and multi-dimensional feature, our framework takes both of the corresponding vertices pair and their global contextual information into consideration, enabling the generated graph representation to have a globally optimal message passing mechanism for different down-stream tasks. The principled investigation results achieved for various graph analysis tasks (e.g., graph classification, vertex classification, and link prediction) on 11 graph and non-graph datasets show that our GRATIS can not only largely enhance pre-defined graphs but also learns a strong graph representation for non-graph data, with clear performance improvements on all tasks. In particular, the learned topology and multi-dimensional edge features provide complementary task-related cues for graph analysis tasks. Our framework is effective, robust and flexible, and is a plug-and-play module that can be combined with different backbones and Graph Neural Networks (GNNs) to generate a task-specific graph representation from various graph and non-graph data. Our code is made publicly available at https://***/SSYSteve/Learning-Graph-Representation-with-Task-specific-Topology-and-Multi-dimensional-Edge-Feature

关键词： Graph 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：