检索结果-内蒙古大学图书馆

Data-free adversarial perturbations for practical black-box attack

学校读者我要写书评

暂无评论

arXiv 2020年

作者： Huan, Zhaoxin Wang, Yulong Zhang, Xiaolu Shang, Lin Fu, Chilin Zhou, Jun Department of Computer Science and Technology State Key Laboratory for Novel Software Technology Nanjing University Ant Financial Services Group Department of Computer Science and Technology Tsinghua university

Neural networks are vulnerable to adversarial examples, which are malicious inputs crafted to fool pre-trained models. Adversarial examples often exhibit black-box attacking transferability, which allows that adversarial examples crafted for one model can fool another model. However, existing black-box attack methods require samples from the training data distribution to improve the transferability of adversarial examples across different models. Because of the data dependence, the fooling ability of adversarial perturbations is only applicable when training data are accessible. In this paper, we present a data-free method for crafting adversarial perturbations that can fool a target model without any knowledge about the training data distribution. In the practical setting of a black-box attack scenario where attackers do not have access to target models and training data, our method achieves high fooling rates on target models and outperforms other universal adversarial perturbation methods. Our method empirically shows that current deep learning models are still at risk even when the attackers do not have access to training data. Copyright © 2020, The Authors. All rights reserved.

关键词： Perturbation techniques

Deep Hyperspectral Unmixing using Transformer Network

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Ghosh, Preetam Roy, Swalpa Kumar Koirala, Bikram Rasti, Behnood Scheunders, Paul The Department of Computer Science and Engineering Jalpaiguri Engineering College West Bengal 735102 India Universiteitsplein 1 AntwerpB-2610 Belgium Helmholtz-Zentrum Dresden-Rossendorf Helmholtz Institute Freiberg for Resource Technology Machine Learning Group Chemnitzer Straße 40 Freiberg 09599 Germany

Transformers have intrigued the vision research community with their state-of-the-art performance in natural language processing. With their superior performance, transformers have found their way in the field of hyperspectral image classification and achieved promising results. In this article, we harness the power of transformers to conquer the task of hyperspectral unmixing and propose a novel deep unmixing model with transformers. We aim to utilize the ability of transformers to better capture the global feature dependencies in order to enhance the quality of the endmember spectra and the abundance maps. The proposed model is a combination of a convolutional autoencoder and a transformer. The hyperspectral data is encoded by the convolutional encoder. The transformer captures long-range dependencies between the representations derived from the encoder. The data are reconstructed using a convolutional decoder. We applied the proposed unmixing model to three widely used unmixing datasets, i.e., Samson, Apex, and Washington DC mall and compared it with the state-of-the-art in terms of root mean squared error and spectral angle distance. The source code for the proposed model will be made publicly available at https://***/preetam22n/DeepTrans-HSU. © 2022, CC BY.

关键词： Convolution

Improved Fine-Tuning by Better Leveraging Pre-Training Data

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Liu, Ziquan Xu, Yi Xu, Yuanhong Qian, Qi Li, Hao Ji, Xiangyang Chan, Antoni B. Jin, Rong Department of Computer Science City University of Hong Kong Hong Kong School of Artificial Intelligence Dalian University of Technology China DAMO Academy Alibaba Group China Department of Automation Tsinghua University China

As a dominant paradigm, fine-tuning a pre-trained model on the target data is widely used in many deep learning applications, especially for small data sets. However, recent studies have empirically shown that training from scratch has the final performance that is no worse than this pre-training strategy once the number of training samples is increased in some vision tasks. In this work, we revisit this phenomenon from the perspective of generalization analysis by using excess risk bound which is popular in learning theory. The result reveals that the excess risk bound may have a weak dependency on the pre-trained model. The observation inspires us to leverage pre-training data for fine-tuning, since this data is also available for fine-tuning. The generalization result of using pre-training data shows that the excess risk bound on a target task can be improved when the appropriate pre-training data is included in fine-tuning. With the theoretical motivation, we propose a novel selection strategy to select a subset from pre-training data to help improve the generalization on the target task. Extensive experimental results for image classification tasks on 8 benchmark data sets verify the effectiveness of the proposed data selection based fine-tuning pipeline. © 2021, CC BY.

关键词： Risk assessment

Integrated Bi-Manual Motion Generation and Control shaped for Probabilistic Movement Primitives

学校读者我要写书评

暂无评论

Integrated Bi-Manual Motion Generation and Control shaped fo...

IEEE-RAS International Conference on Humanoid Robots

作者： Jonathan Vorndamme João Carvalho Riddhiman Laha Dorothea Koert Luis Figueredo Jan Peters Sami Haddadin Department of Computer Engineering Technische Universität München (TUM) TUM School of Computation Information and Technology (CIT) Munich Institute of Robotics and Machine Intelligence (MIRMI) Germany Intelligent Autonomous Systems group Technische Universitat Darmstadt (TUD) Germany Center for Cognitive Science TUD Germany

ISBN: (纸本)9798350309805

This work introduces a novel cooperative control framework that allows for real-time reactiveness and adaptation whilst satisfying implicit constraints stemming from proba-bilistic/stochastic trajectories. Stemming from task-oriented sampling and/or task-oriented demonstrations, e.g., learning based on motion primitives, such trajectories carry additional information often neglected during real-time control deployment. In particular, methods such as probabilistic movement primitives offer the advantage to capture the inherent stochasticity in human demonstrations - which in turn reflects human's understanding about task-variability and adaption possibilities. This information, however, is often poorly exploited and, mostly, used during offline trajectory planning stage. Our work instead introduces a novel real-time motion-generation strategy that explicitly exploits such information to improve trajectories according to changes in the environmental condition and robot task-space topology. The proposed solution is particularly well-suited for bi-manual and coordinated systems where the increased kinematic complexity, tightly-coupled constraints and reduced workspace have detrimental effects on the manipula-bility, joint-limits, and are even capable of causing unstable behavior and task-failure. Our methodology addresses these challenges, and improves performance and task-execution by taking the confidence range region explicitly into account whilst maneuvering towards better configurations. Furthermore, it can directly cope with different closed-chain kinematics and task-space topologies, resulting for instance from different grasps. Experimental evaluations on a bi-manual Franka panda robot show that the method can run in the inner control loop of the robot and enables successful execution of highly constrained tasks.

关键词： Trajectory planning Robot kinematics Humanoid robots Kinematics Probabilistic logic Real-time systems Trajectory

SMILE-UHURA Challenge - Small Vessel Segmentation at Mesoscopic Scale from Ultra-High Resolution 7T Magnetic Resonance Angiograms

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chatterjee, Soumick Mattern, Hendrik Dörner, Marc Sciarra, Alessandro Dubost, Florian Schnurre, Hannes Khatun, Rupali Yu, Chun-Chih Hsieh, Tsung-Lin Tsai, Yi-Shan Fang, Yi-Zeng Yang, Yung-Ching Huang, Juinn-Dar Xu, Marshall Liu, Siyu Ribeiro, Fernanda L. Bollmann, Saskia Chintalapati, Karthikesh Varma Radhakrishna, Chethan Mysuru Ram Kumar, Sri Chandana Hudukula Sutrave, Raviteja Qayyum, Abdul Mazher, Moona Razzak, Imran Rodero, Cristobal Niederer, Steven Lin, Fengming Xia, Yan Wang, Jiacheng Qiu, Riyu Wang, Liansheng Panah, Arya Yazdan Jurdi, Rosana El Fu, Guanghui Arslan, Janan Vaillant, Ghislain Valabregue, Romain Dormont, Didier Stankoff, Bruno Colliot, Olivier Vargas, Luisa Chacón, Isai Daniel Pitsiorlas, Ioannis Arbeláez, Pablo Zuluaga, Maria A. Schreiber, Stefanie Speck, Oliver Nürnberger, Andreas Faculty of Computer Science Otto von Guericke University Magdeburg Magdeburg Germany Data and Knowledge Engineering Group Otto von Guericke University Magdeburg Magdeburg Germany Genomics Research Centre Human Technopole Milan Italy Biomedical Magnetic Resonance Otto von Guericke University Magdeburg Magdeburg Germany Department of Neurology Medical Faculty University Hospital of Magdeburg Magdeburg Germany German Centre for Neurodegenerative Diseases Magdeburg Germany Centre for Behavioural Brain Sciences Magdeburg Germany University Hospital Zurich University of Zurich Zurich Switzerland Google Inc. United States Translational Radiobiology Department of Radiation Oncology Universitätsklinikum Erlangen Friedrich-Alexnder-Universität Erlangen-Nürnberg Erlangen Germany National Yang Ming Chiao Tung University Hsinchu Taiwan School of Electrical Engineering and Computer Science University of Queensland Brisbane Australia Australian eHealth Research Centre Chile National Heart and Lung Institute Faculty of Medicine Imperial College London London United Kingdom Centre for Medical Image Computing Department of Computer Science University College London London United Kingdom School of Computer Science and Engineering University of New South Wales Sydney Australia Mohamed bin Zayed University of Artificial Intelligence Abu Dhabi United Arab Emirates The Alan Turing Institute London United Kingdom School of Computing University of Leeds Leeds United Kingdom Department of Computer Science School of Informatics Xiamen University Xiamen China Manteia Technologies Co. Ltd Xiamen China Leicester International Institute Dalian University of Technology Dalian China Sorbonne Université Institut du Cerveau - Paris Brain Institute ICM CNRS Inria Inserm AP-HP Hôpital de la Pitié Salpêtrière ParisF-75013 France Centre of Formation and Research in Artificial Intelligence Universidad de Los Andes Colombia Data Science Department EURECOM Sophia Antipolis Franc

The human brain receives nutrients and oxygen through an intricate network of blood vessels. Pathology affecting small vessels, at the mesoscopic scale, represents a critical vulnerability within the cerebral blood supply and can lead to severe conditions, such as Cerebral Small Vessel Diseases. The advent of 7 Tesla MRI systems has enabled the acquisition of higher spatial resolution images, making it possible to visualise such vessels in the brain. However, the lack of publicly available annotated datasets has impeded the development of robust, machine learning-driven segmentation algorithms. To address the complexities of mesoscopic vessel segmentation and to highlight the need for advanced techniques to manage the high noise levels and poor vessel-to-background contrast inherent in"ultra-high-resolution" data, the SMILE-UHURA challenge was organised. This challenge, held in conjunction with the ISBI 2023, in Cartagena de Indias, Colombia, aimed to provide a platform for researchers working on related topics. The SMILE-UHURA challenge addresses the gap in publicly available annotated datasets by providing an annotated dataset of Time-of-Flight angiography acquired with 7T MRI. This dataset was created through a combination of automated pre-segmentation and extensive manual refinement. In this manuscript, sixteen submitted methods and two baseline methods are compared both quantitatively and qualitatively on two different datasets: held-out test MRAs from the same dataset as the training data (with labels kept secret) and a separate 7T ToF MRA dataset where both input volumes and labels are kept secret. The results demonstrate that most of the submitted deep learning methods, trained on the provided training dataset, achieved reliable segmentation performance. Dice scores reached up to 0.838 ± 0.066 and 0.716 ± 0.125 on the respective datasets, with an average performance of up to 0.804 ± 0.15. Copyright © 2024, The Authors. All rights reserved.

关键词： Deep learning

Satellite-based InSAR Geodesy and Collocation with GNSS

学校读者我要写书评

暂无评论

Satellite-based InSAR Geodesy and Collocation with GNSS

2023 International Conference on ENTERprise Information Systems, CENTERIS 2023 - International Conference on Project MANagement, ProjMAN 2023 - International Conference on Health and Social Care Information Systems and Technologies, HCist 2023

作者： Papco, Juraj Bakon, Matus Kubica, Lukas Belicova, Gabika Droscak, Branislav Ferianc, Martin Rovnak, Martin Ruiz, Antonio M. Sousa, Joaquim J. Department of Theoretical Geodesy and Geoinformatics Faculty of Civil Engineering Slovak University of Technology in Bratislava Radlinskeho 11 Bratislava813 62 Slovakia Insar.sk Ltd Konstantinova 3 Presov080 01 Slovakia Department of Finance Accounting and Mathematical Methods Faculty of Management and Business University of Presov Konstantinova 16 Presov080 01 Slovakia Department of Engineering School of Sciences and Technology Universidade de Tras-os-Montes e Alto Douro Quinta de Prados Vila Real5000-801 Portugal Geodetic and Cartographic Institute Bratislava Chlumeckeho 4 Bratislava827 45 Slovakia Department of Cartographic Geodetic and Photogrammetry Engineering University of Jaen Campus Las Lagunillas s/n Jaen23071 Spain University of Jaen Campus Las Lagunillas s/n Jaen23071 Spain Research Group RNM-282 Microgeodesia Jaen University of Jaen Campus Las Lagunillas s/n Jaen23071 Spain Institute for Systems and Computer Engineering Technology and Science INESC-TEC Porto4200-465 Portugal

Satellite radar interferometry (InSAR) enables large-scale monitoring of displacement with precision akin to ground-based methods. Using measurements from Sentinel-1 satellites, persistent scatterer networks cover areas up to hundreds of kilometers in size. One of the current challenges for the InSAR community is to tie these networks into a current and country-specific global terrestrial reference system through collocation with other geodetic techniques, especially Global Navigation Satellite Systems (GNSS). In this work, an optimal procedure for assessing a suitable collocation properties for InSAR and GNSS techniques is discussed, along with the proposal for the methodology to determine the spatial relationship between the reference points of both techniques. The study analyzes three collocation stations in Slovakia using their radar backscattering characteristics. InSAR achieves an intrinsic accuracy of up to 1 mm but has uncertainties several orders of magnitude larger in 3D space. A collocation of InSAR measurements with GNSS can tie relative InSAR displacement estimates to an absolute reference frame. © 2024 The Author(s). Published by Elsevier B.V.

关键词： Geodesy

Fast and Accurate Fault Detection and Classification in Transmission Lines using Extreme Learning Machine

学校读者我要写书评

暂无评论

e-Prime - Advances in Electrical Engineering, Electronics and Energy 2023年 3卷

作者： Goni, Md.Omaer Faruq Nahiduzzaman, Md. Anower, Md.Shamim Rahman, Md.Mahabubur Islam, Md.Robiul Ahsan, Mominul Haider, Julfikar Shahjalal, Mohammad Department of Electrical & Computer Engineering Rajshahi University of Engineering & Technology Rajshahi 6204 Bangladesh Department of Electrical & Electronic Engineering Rajshahi University of Engineering & Technology Rajshahi 6204 Bangladesh Department of Computer Science University of York Deramore Lane NY Heslington YO10 5GH Ireland Department of Engineering Manchester Metropolitan University Chester St Manchester M1 5GD Ireland WMG Warwick Manufacturing Group University of Warwick Coventry CV4 7AL United Kingdom Department of Electrical Engineering Qatar University Doha 2713 Qatar

To provide stability and a continuous supply of power, the detection and classification of faults in the transmission lines (TLs) are crucial in this modern age. It is required to remove a faulty section from a healthy section to provide safety and to minimize power loss due to the fault. In the contemporary world, machine learning (ML) is extensively used in every aspect of life. In this study, a spontaneous fault detection (FD) and fault classification (FC) system based on ML has been proposed. MATLAB Simulink was employed to simulate two different TLs and to generate normal and fault data (Per unit voltage and current) of ten different types. TL-1 consisted of a single generator and a single load whereas TL-2 consisted of two generators and three loads. Upon normalizing the data, an extreme learning machine (ELM) algorithm was used as the classifier. Two different ELM models were developed for FD and FC purposes through training. The method achieved fault classification accuracies of 99.18% and 99.09% for the TL-1 and TL-2 respectively. On the other hand, fault detection accuracies of 99.53% and 99.60% were achieved for the TL-1 and TL-2. The proposed ELM model compared to a traditional artificial neural network (ANN) model demonstrated relatively a shorter processing time and reduced computational complexity. In addition, the proposed method outperformed the existing state-of-the-art methods. © 2023 The Author(s)

关键词： Extreme learning machine Fault classification Fault detection Matlab simulink Transmission line

Correction: Multi-modal characterisation of early-stage, subclinical cardiac deterioration in patients with type 2 diabetes

学校读者我要写书评

暂无评论

Cardiovascular diabetology 2025年第1期24卷 43页

作者： Ambre Bertrand Andrew Lewis Julia Camps Vicente Grau Blanca Rodriguez Computational Cardiovascular Science Group Department of Computer Science University of Oxford Oxford OX1 3QD UK. ambre.bertrand@cs.ox.ac.uk. Division of Cardiovascular Medicine Radcliffe Department of Medicine University of Oxford Oxford OX3 9DU UK. Computational Cardiovascular Science Group Department of Computer Science University of Oxford Oxford OX1 3QD UK. Institute of Biomedical Engineering Department of Engineering Science University of Oxford Oxford OX3 7DQ UK. Computational Cardiovascular Science Group Department of Computer Science University of Oxford Oxford OX1 3QD UK. blanca.rodriguez@cs.ox.ac.uk.

Prediction of micrometeoroid damage to lunar construction materials using numerical modeling of hypervelocity impact events 15

学校读者我要写书评

暂无评论

Prediction of micrometeoroid damage to lunar construction ma...

2019 15th Hypervelocity Impact Symposium, HVIS 2019

作者： Allende, Maria I. Miller, Joshua E. Davis, B. Alan Christiansen, Eric L. Lepech, Michael D. Loftus, David J. Department of Civil and Environmental Engineering Stanford University StanfordCA94305 United States Department of Computer Science University of Texas at El Paso El PasoTX79968 United States Jacobs Technologies Engineering Science Contract Group NASA Johnson Space Center HoustonTX77058 United States Astromaterials Research and Exploration Science NASA Johnson Space Center HoustonTX77058 United States Division of Space Biosciences NASA Ames Research Center Moffett FieldCA94035 United States