检索结果-内蒙古大学图书馆

To Make Yourself Invisible with Adversarial Semantic Contours

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Zhang, Yichi Zhu, Zijian Su, Hang Zhu, Jun Zheng, Shibao He, Yuan Xue, Hui Department of Computer Science and Technology Institute for Artificial Intelligence THBI Lab Tsinghua University Beijing100084 China Institute of Image Communication and Network Engineering Shanghai Jiao Tong University Shanghai 200240 China Alibaba Group Hangzhou311121 China

Modern object detectors are vulnerable to adversarial examples, which may bring risks to real-world applications. The sparse attack is an important task which, compared with the popular adversarial perturbation on the whole image, needs to select the potential pixels that is generally regularized by an l0-norm constraint, and simultaneously optimize the corresponding texture. The non-differentiability of l0 norm brings challenges and many works on attacking object detection adopted manually-designed patterns to address them, which are meaningless and independent of objects, and therefore lead to relatively poor attack performance. In this paper, we propose Adversarial Semantic Contour (ASC), an MAP estimate of a Bayesian formulation of sparse attack with a deceived prior of object contour. The object contour prior effectively reduces the search space of pixel selection and improves the attack by introducing more semantic bias. Extensive experiments demonstrate that ASC can corrupt the prediction of 9 modern detectors with different architectures (\eg, one-stage, two-stage and Transformer) by modifying fewer than 5% of the pixels of the object area in COCO in white-box scenario and around 10% of those in black-box scenario. We further extend the attack to datasets for autonomous driving systems to verify the effectiveness. We conclude with cautions about contour being the common weakness of object detectors with various architecture and the care needed in applying them in safety-sensitive scenarios. © 2022, The Authors. All rights reserved.

关键词： Semantics

1D quasi-bound states in the continuum with large operation bandwidth in theω~k space for nonlinear optical applications

学校读者我要写书评

暂无评论

Photonics Research 2022年第7期10卷 1575-1581页

作者： KAILI SUN HUI JIANG DMITRY ABYKOV VIEN VAN URIEL LEVY YANGJIAN CAI ZHANGHUA HAN Shandong Provincial Key Laboratory of Optics and Photonic Devices Center of Light Manipulation and ApplicationsSchool of Physics and ElectronicsShandong Normal UniversityJinan 250358China Image Processing Systems Institute—Branch of the Federal Scientific Research Centre“Crystallography and Photonics”of Russian Academy of Sciences Samara 443001Russia Department of Electrical and Computer Engineering University of AlbertaEdmontonAlberta T6G 2V4Canada Department of Applied Physics and Center for Nanoscience and NanotechnologyThe Hebrew University of JerusalemJerusalemIsrael

The phenomenon of bound state in the continuum(BIC)with an infinite quality factor and lifetime has emerged in recent years in photonics as a new tool to manipulate light-matter ***,most of the investigated structures only support BIC resonances at very few discrete points in theω~k *** when the BIC is switched to a quasi-BIC(QBIC)resonance through perturbation,its frequency will still be located within a narrow spectral band close to that of the original BIC,restricting their applications in many fields where random or multiple input frequencies beyond the narrow band are *** this work,we demonstrate that a new set of QBIC resonances can be supported by using a special binary grating consisting of two alternatingly aligned ridge arrays with the same period and zero-approaching ridge width difference on a slab *** QBIC resonances are distributed continuously over a broad band along a line in theω~k space and can thus be considered as 1 D *** the Q factors generally affected by the ridge difference,it is now possible to arbitrarily choose any frequencies on the dispersion line to achieve significantly enhanced light-matter interactions,facilitating many applications where multiple input wavelengths are required;e.g.,sum or difference frequency generations in nonlinear optics.

关键词： quasi nonlinear narrow

SynCLay: Interactive Synthesis of Histology images from Bespoke Cellular Layouts

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Deshpande, Srijay Dawood, Muhammad Minhas, Fayyaz Rajpoot, Nasir Tissue Image Analytics Centre Department of Computer Science University of Warwick United Kingdom The Alan Turing Institute London United Kingdom Department of Pathology University Hospitals Coventry & Warwickshire United Kingdom Histofy Ltd Birmingham United Kingdom

Automated synthesis of histology images has several potential applications in computational pathology. However, no existing method can generate realistic tissue images with a bespoke cellular layout or user-defined histology parameters. In this work, we propose a novel framework called SynCLay (Synthesis from Cellular Layouts) that can construct realistic and high-quality histology images from user-defined cellular layouts along with annotated cellular boundaries. Tissue image generation based on bespoke cellular layouts through the proposed framework allows users to generate different histological patterns from arbitrary topological arrangement of different types of cells (e.g., neutrophils, lymphocytes, epithelial cells and others). SynCLay generated synthetic images can be helpful in studying the role of different types of cells present in the tumor microenvironmet. Additionally, they can assist in balancing the distribution of cellular counts in tissue images for designing accurate cellular composition predictors by minimizing the effects of data imbalance. We train SynCLay in an adversarial manner and integrate a nuclear segmentation and classification model in its training to refine nuclear structures and generate nuclear masks in conjunction with synthetic images. During inference, we combine the model with another parametric model for generating colon images and associated cellular counts as annotations given the grade of differentiation and cellularities (cell densities) of different cells. We assess the generated images quantitatively using the Frechet Inception Distance and report on feedback from trained pathologists who assigned realism scores to a set of images generated by the framework. The average realism score across all pathologists for synthetic images was as high as that for the real images. We demonstrate that the proposed framework can be used to add new cells to a tissue images and alter cellular positions. We also show that augmenting limite

关键词： Cells

Normative Diffusion Autoencoders: Application to Amyotrophic Lateral Sclerosis

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Ijishakin, Ayodeji Hadjasavilou, Adamos Abdulaal, Ahmed Montana-Brown, Nina Townend, Florence Spinelli, Edoardo Filippi, Massimo Agosta, Federica Cole, James Malaspina, Andrea Centre for Medical Image Computing Department of Computer Science University College London United Kingdom Department of Neuromuscular Diseases University College London United Kingdom Neuroimaging Research Unit Division of Neuroscience IRCCS San Raffaele Scientific Institute Milan Italy Vita-Salute San Raffaele University Milan Italy Neurology Unit IRCCS San Raffaele Scientific Institute Milan Italy Neurorehabilitation Unit IRCCS San Raffaele Scientific Institute Milan Italy Neurophysiology Service IRCCS San Raffaele Scientific Institute Milan Italy

Predicting survival in Amyotrophic Lateral Sclerosis (ALS) is a challenging task. Magnetic resonance imaging (MRI) data provide in vivo insight into brain health, but the low prevalence of the condition and resultant data scarcity limit training set sizes for prediction models. Survival models are further hindered by the subtle and often highly localised profile of ALS-related neurodegeneration. Normative models present a solution as they increase statistical power by leveraging large healthy cohorts. Separately, diffusion models excel in capturing the semantics embedded within images including subtle signs of accelerated brain ageing, which may help predict survival in ALS. Here, we combine the benefits of generative and normative modelling by introducing the normative diffusion autoencoder framework. To our knowledge, this is the first use of normative modelling within a diffusion autoencoder, as well as the first application of normative modelling to ALS. Our approach outperforms generative and non-generative normative modelling benchmarks in ALS prognostication, demonstrating enhanced predictive accuracy in the context of ALS survival prediction and normative modelling in general. © 2024, CC BY-NC-SA.

关键词： Diffusion

Multi-Task Diffusion Incentive Design for Mobile Crowdsourcing in Social Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Guo, Jianxiong Ni, Qiufen Wu, Weili Du, Ding-Zhu The Advanced Institute of Natural Sciences Beijing Normal University Zhuhai519087 China The Guangdong Key Lab of AI and Multi-Modal Data Processing BNU-HKBU United International College Zhuhai519087 China The School of Computers Guangdong University of Technology Guangzhou510006 China The Department of Computer Science Erik Jonsson School of Engineering and Computer Science The University of Texas at Dallas RichardsonTX75080 United States

Mobile Crowdsourcing (MCS) is a novel distributed computing paradigm that recruits skilled workers to perform location-dependent tasks. A number of mature incentive mechanisms have been proposed to address the worker recruitment problem in MCS systems. However, they all assume that there is a large enough worker pool and a sufficient number of users can be selected. This may be impossible in large-scale crowdsourcing environments. To address this challenge, we consider the MCS system defined on a location-aware social network provided by a social platform. In this system, we can recruit a small number of seed workers from the existing worker pool to spread the information of multiple tasks in the social network, thus attracting more users to perform tasks. In this paper, we propose a Multi-Task Diffusion Maximization (MT-DM) problem that aims to maximize the total utility of performing multiple crowdsourcing tasks under the budget. To accommodate multiple tasks diffusion over a social network, we create a multi-task diffusion model, and based on this model, we design an auction-based incentive mechanism, MT-DM-L. To deal with the high complexity of computing the multi-task diffusion, we adopt Multi-Task Reverse Reachable (MT-RR) sets to approximate the utility of information diffusion efficiently. Through both complete theoretical analysis and extensive simulations by using real-world datasets, we validate that our estimation for the spread of multitask diffusion is accurate and the proposed mechanism achieves individual rationality, truthfulness, computational efficiency, and (1 − 1/√e − Ε) approximation with at least 1 − δ probability. Copyright © 2023, The Authors. All rights reserved.

关键词： Approximation algorithms

Controlling Unknown Quantum States via Data-Driven State Representations

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhu, Yan Xiao, Tailong Zeng, Guihua Chiribella, Giulio Wu, Ya-Dong QICI Quantum Information and Computation Initiative Department of Computer Science The University of Hong Kong Pokfulam Road Hong Kong State Key Laboratory of Advanced Optical Communication Systems and Networks Institute for Quantum Sensing and Information Processing Shanghai Jiao Tong University Shanghai200240 China Hefei National Laboratory Hefei230088 China Shanghai Research Center for Quantum Sciences Shanghai201315 China Department of Computer Science Parks Road OxfordOX1 3QD United Kingdom Perimeter Institute for Theoretical Physics WaterlooONN2L 2Y5 Canada John Hopcroft Center for Computer Science Shanghai Jiao Tong University Shanghai200240 China

Accurate control of quantum states is crucial for quantum computing and other quantum technologies. In the basic scenario, the task is to steer a quantum system towards a target state through a sequence of control operations. Determining the appropriate operations, however, generally requires information about the initial state of the system. Gathering this information becomes increasingly challenging when the initial state is not a priori known and the system’s size grows large. To address this problem, we develop a machine-learning algorithm that uses a small amount of measurement data to construct its internal representation of the system’s state. The algorithm compares this data-driven representation with a representation of the target state, and uses reinforcement learning to output the appropriate control operations. We illustrate the effectiveness of the algorithm showing that it achieves accurate control of unknown many-body quantum states and non-Gaussian continuous-variable states using data from a limited set of quantum measurements. © 2024, CC BY-NC-SA.

关键词： Reinforcement learning

VAN: Voting and Attention Based Network for Unsupervised Medical image Registration 18th

学校读者我要写书评

暂无评论

VAN: Voting and Attention Based Network for Unsupervised Med...

18th Pacific Rim International Conference on Artificial Intelligence, PRICAI 2021

作者： Zu, Zhiang Zhang, Guixu Peng, Yaxin Ye, Zhen Shen, Chaomin School of Computer Science East China Normal University Shanghai China Department of Mathematics Shanghai University Shanghai China Shanghai Electric Central Research Institute Shanghai China Shanghai Key Laboratory of Multidimensional Information Processing East China Normal University Shanghai China

ISBN: (纸本)9783030891879

In this paper, a novel unsupervised network for medical image registration called VAN (Voting and Attention based Network) is proposed, in which the final deformation field is determined by the voting process between multiple registration branches. To reduce model parameters, multiple registration branches share one encoder. Besides, the attention mechanism is introduced, which further improves the network accuracy. We also adopt the method of single training and multiple-registrations to deal with the problem of the large deformation field. The experimental results show that the registration effect of our proposed network outperforms the baselines VoxelMorph and Symmetric Normalization (SyN) on three brain MRI image datasets. © 2021, Springer Nature Switzerland AG.

关键词： Deformation

A Gröbner Basis Approach to Combinatorial Nullstellensatz

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Xu, Yang Kan, Haibin Han, Guangyue Department of Mathematics Faculty of Science The University of Hong Kong Pokfulam Road Hong Kong Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China Shanghai Engineering Research Center of Blockchain Shanghai200433 China Yiwu Research Institute Fudan University Zhejiang Yiwu City322000 China

In this paper, using some conditions that arise naturally in Alon's combinatorial Nullstellensatz as well as its various extensions and generalizations, we characterize Gröbner bases consisting of monic polynomials, which helps us to establish a Nullstellensatz from a Gröbner basis perspective. As corollaries of this general Nullstellensatz, we establish four special Nullstellensatz, which, among others, include a common generalization of the Nullstellensatz for multisets established in Kós, Rónyai and Mészáros [23, 24] and the Nullstellensatz with multiplicity established in Ball and Serra [9], and include a punctured Nullstellensatz, generalizing several existing results in the literature. As applications of our punctured Nullstellensatz, we extend some results on hyperplane covering in [9, 23, 24] to wider settings, and give an alternative proof of the generalized Alon-Füredi theorem established in Bishnoi, Clark, Potukuchi and Schmitt [12]. Unless specified otherwise, all our results are established over an arbitrary commutative ring R. © 2023, CC BY.

关键词：

Modification of a Herbal MRI Contrast Agent: Biological Effect and Relaxivity

学校读者我要写书评

暂无评论

Modification of a Herbal MRI Contrast Agent: Biological Effe...

Iranian Conference of Biomedical Engineering (ICBME)

作者： Ali Yazdani Ahmadreza Okhovat Raheleh Doosti Maryam Saber Hamid Soltanian-Zadeh Control and Intelligent Processing Center of Excellence School of Electrical and Computer Engineering University of Tehran Tehran Iran Central Herbarium School of Biology College of Science University of Tehran Tehran Iran College of Science University of Tehran Tehran Iran Department of Stem Cells and Developmental Biology Royan Institute for Stem Cell Biology and Technology Tehran Iran

We modified a modern herbal MRI contrast agent for biological applications. This modification was realized by omitting salinity and free ions of the contrast media. As a result, both the toxicity and the relaxivity of the product improved. We also chemically analyzed the modified agent and confirmed the elimination of some annoying free ions and molecules.

关键词：