检索结果-内蒙古大学图书馆

Frequency-driven Imperceptible Adversarial Attack on Semantic Similarity

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Luo, Cheng Lin, Qinliang Xie, Weicheng Wu, Bizhu Xie, Jinheng Shen, Linlin Computer Vision Institute School of Computer Science & Software Engineering Shenzhen University2Shenzhen Institute of Artificial Intelligence & Robotics for Society3Guangdong Key Laboratory of Intelligent Information Processing

Current adversarial attack research reveals the vulnerability of learning-based classifiers against carefully crafted perturbations. However, most existing attack methods have inherent limitations in cross-dataset generalization as they rely on a classification layer with a closed set of categories. Furthermore, the perturbations generated by these methods may appear in regions easily perceptible to the human visual system (HVS). To circumvent the former problem, we propose a novel algorithm that attacks semantic similarity on feature representations. In this way, we are able to fool classifiers without limiting attacks to a specific dataset. For imperceptibility, we introduce the low-frequency constraint to limit perturbations within high-frequency components, ensuring perceptual similarity between adversarial examples and originals. Extensive experiments on three datasets (CIFAR-10, CIFAR-100, and ImageNet-1K) and three public online platforms indicate that our attack can yield misleading and transferable adversarial examples across architectures and datasets. Additionally, visualization results and quantitative performance (in terms of four different metrics) show that the proposed algorithm generates more imperceptible perturbations than the state-of-the-art methods. Code is made available at https://***/LinQinLiang/SSAH-adversarial-attack. © 2022, CC BY.

关键词： Classification (of information)

Rearranging the Environment to Maximize Energy with a Robotic Circuit Drawing

学校读者我要写书评

暂无评论

Rearranging the Environment to Maximize Energy with a Roboti...

IEEE International Conference on robotics and Biomimetics

作者： Xianglong Tan Zhikang Liu Chen Yu Sören Schwertfeger Andre Rosendo Hamlyn Centre Imperial College London UK Department of Computer Science University of York UK School of Information Science and Technology ShanghaiTech University China Robotics Engineering Worcester Polytechnic Institute USA

ISBN: (纸本)9781665481106

The ability of acquiring energy form environments is greatly beneficial for robot to function in uncertain environments. In this work, we present a robot capable of drawing circuits with conductive ink while also rearranging the visual world to receive maximum energy from a power source. A range of circuit drawing tasks is designed to simulate real-world scenarios, including avoiding physical obstacles and regions that would discontinue drawn circuits. We adopt the state-of-the-art Transporter networks for pick-and-place manipulation from visual observation. We conduct experiments in both simulation and real-world settings, and our results show that, with a small number of demonstrations, the robot learns to rearrange the placement of objects (removing obstacles and bridging areas unsuitable for drawing) and to connect a power source with a minimum amount of conductive ink. As autonomous robots become more and more involved in our daily lives, our proposed approach brings a novel way for machines to keep themselves functional by rearranging their surroundings to create their own electric circuits.

关键词： Training Visualization Power supplies Planets Surface resistance Ink Machine learning

ENTROPY IS NOT ENOUGH FOR TEST-TIME ADAPTATION: FROM THE PERSPECTIVE OF DISENTANGLED FACTORS

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lee, Jonghyun Jung, Dahuin Lee, Saehyung Park, Junsung Shin, Juhyeon Hwang, Uiwon Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of School of Computer Science and Engineering Soongsil University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Division of Digital Healthcare Yonsei University Korea Republic of AIIS ASRI INMC and ISRC Seoul National University Korea Republic of

Test-time adaptation (TTA) fine-tunes pre-trained deep neural networks for unseen test data. The primary challenge of TTA is limited access to the entire test dataset during online updates, causing error accumulation. To mitigate it, TTA methods have utilized the model output’s entropy as a confidence metric that aims to determine which samples have a lower likelihood of causing error. Through experimental studies, however, we observed the unreliability of entropy as a confidence metric for TTA under biased scenarios and theoretically revealed that it stems from the neglect of the influence of latent disentangled factors of data on predictions. Building upon these findings, we introduce a novel TTA method named Destroy Your Object (DeYO), which leverages a newly proposed confidence metric named Pseudo-Label Probability Difference (PLPD). PLPD quantifies the influence of the shape of an object on prediction by measuring the difference between predictions before and after applying an object-destructive transformation. DeYO consists of sample selection and sample weighting, which employ entropy and PLPD concurrently. For robust adaptation, DeYO prioritizes samples that dominantly incorporate shape information when making predictions. Our extensive experiments demonstrate the consistent superiority of DeYO over baseline methods across various scenarios, including biased and wild. Project page is publicly available at https://***/DeYO/. Copyright © 2024, The Authors. All rights reserved.

关键词： Entropy

EFFICIENT UNIVERSAL SHUFFLE ATTACK FOR VISUAL OBJECT TRACKING

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Liu, Siao Chen, Zhaoyu Li, Wei Zhu, Jiwei Wang, Jiafeng Zhang, Wenqiang Gan, Zhongxue Academy for Engineering and Technology Fudan University Shanghai China School of Computer Science Fudan University Shanghai China The Department of Engineering Research Center for Intelligent Robotics Ji Hua Laboratory Foshan China

Recently, adversarial attacks have been applied in visual object tracking to deceive deep trackers by injecting imperceptible perturbations into video frames. However, previous work only generates the video-specific perturbations, which restricts its application scenarios. In addition, existing attacks are difficult to implement in reality due to the real-time of tracking and the re-initialization mechanism. To address these issues, we propose an offline universal adversarial attack called Efficient Universal Shuffle Attack. It takes only one perturbation to cause the tracker malfunction on all videos. To improve the computational efficiency and attack performance, we propose a greedy gradient strategy and a triple loss to efficiently capture and attack model-specific feature representations through the gradients. Experimental results show that EUSA can significantly reduce the performance of state-of-the-art trackers on OTB2015 and VOT2018. Copyright © 2022, The Authors. All rights reserved.

关键词： Computational efficiency

Q-learning based Path Planning Method for UAVs using Priority Shifting

学校读者我要写书评

暂无评论

Q-learning based Path Planning Method for UAVs using Priorit...

International Conference on Unmanned Aircraft Systems (ICUAS)

作者： Kevin B. de Carvalho Iure Rosa L. de Oliveira Daniel K. D. Villa Alexandre G. Caldeira Má rio Sarcinelli-Filho Alexandre S. Brandã o Department of Electrical Engineering Graduate Program in Computer Science Nucleus of Specialization in Robotics Universidade Federal de Vi&#x00E7 osa Vi&#x00E7 osa MG Brazil Department of Electrical Engineering Graduate Program in Electrical Engineering Universidade Federal do Esp&#x00ED rito Santo Vit&#x00F3 ria ES Brazil

ISBN: (数字)9781665405935

ISBN: (纸本)9781665405942

Path planning is a crucial part of autonomous navigation when regarding autonomous aerial vehicles, often demanding different priorities such as the length, safety or energy consumption. Dynamic programming and geometric methods have been applied to solve this problem, but in recent years, more work has been developed using artificial intelligence approaches, such as reinforcement learning. In this paper we propose an offline path planning method for static environments using Q-learning. An optimal policy is found weighting three important factors: path length, safety and energy consumption. Due to a well balanced exploring/exploiting ratio, the proposed method can lead the agent to the desired destination starting from anywhere in the map. Simulations are done in different scenarios to address the performance of the proposed method and it showcased that the algorithm is able to find feasible paths in each scenario while regarding different set of priorities.

关键词： Energy consumption Q-learning Atmospheric modeling Autonomous aerial vehicles Path planning Safety Dynamic programming

Real-Time Human Action Recognition on Embedded Platforms

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Wang, Ruiqi Wang, Zichen Gao, Peiqi Li, Mingzhen Jeong, Jaehwan Xu, Yihang Lee, Yejin Baum, Carolyn M. Connor, Lisa Tabor Lu, Chenyang AI for Health Institute Washington University St. Louis United States Department of Computer Science and Engineering Washington University St. Louis United States Program in Occupational Therapy Washington University School of Medicine United States

With advancements in computer vision and deep learning, video-based human action recognition (HAR) has become practical. However, due to the complexity of the computation pipeline, running HAR on live video streams incurs excessive delays on embedded platforms. This work tackles the real-time performance challenges of HAR with four contributions: 1) an experimental study identifying a standard Optical Flow (OF) extraction technique as the latency bottleneck in a state-of-the-art HAR pipeline, 2) an exploration of the latency-accuracy tradeoff between the standard and deep learning approaches to OF extraction, which highlights the need for a novel, efficient motion feature extractor, 3) the design of Integrated Motion Feature Extractor (IMFE), a novel single-shot neural network architecture for motion feature extraction with drastic improvement in latency, 4) the development of RT-HARE, a real-time HAR system tailored for embedded platforms. Experimental results on an Nvidia Jetson Xavier NX platform demonstrated that RT-HARE realizes real-time HAR at a video frame rate of 30 frames per second while delivering high levels of recognition accuracy. Copyright © 2024, The Authors. All rights reserved.

关键词： Optical flows

Unbiased Approximations for Stationary Distributions of McKean-Vlasov SDEs

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Awadelkarim, Elsiddig Chada, Neil K. Jasra, Ajay Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia Department of Mathematics City University of Hong Kong Hong Kong School of Data Science The Chinese University of Hong Kong Shenzhen Shenzhen China

We consider the development of unbiased estimators, to approximate the stationary distribution of Mckean-Vlasov stochastic differential equations (MVSDEs). These are an important class of processes, which frequently appear in applications such as mathematical finance, biology and opinion dynamics. Typically the stationary distribution is unknown and indeed one cannot simulate such processes exactly. As a result one commonly requires a time-discretization scheme which results in a discretization bias and a bias from not being able to simulate the associated stationary distribution. To overcome this bias, we present a new unbiased estimator taking motivation from the literature on unbiased Monte Carlo. We prove the unbiasedness of our estimator, under assumptions. In order to prove this we require developing ergodicity results of various discrete time processes, through an appropriate discretization scheme, towards the invariant measure. Numerous numerical experiments are provided, on a range of MVSDEs, to demonstrate the effectiveness of our unbiased estimator. Such examples include the Currie-Weiss model, a 3D neuroscience model and a parameter estimation problem. © 2024, CC BY-NC-SA.

关键词： Stochastic systems

Drone Application Model for Image Acquisition of Plantation Areas and Oil Palm Trees Counting 7

学校读者我要写书评

暂无评论

Drone Application Model for Image Acquisition of Plantation ...

7th International Conference on Information Management and Technology, ICIMTech 2022

作者： Sastrohartono, Hermantoro Suryotomo, Andiko Putro Saifullah, Shoffan Suparyanto, Teddy Perbangsa, Anzaludin Samsinga Pardamean, Bens Instiper Yogyakarta Faculty of Agricultural Technology Department of Agricultural Engineering Yogyakarta55281 Indonesia Universitas Pembangunan Nasional Veteran Yogyakarta Department of Informatics Yogyakarta55283 Indonesia Bioinformatics and Data Science Research Center Bina Nusantara University Jakarta11480 Indonesia School of Information Systems Bina Nusantara University Information Systems Department Jakarta11480 Indonesia BINUS Graduate Program - Master of Computer Science Bina Nusantara University Computer Science Department Jakarta11480 Indonesia

ISBN: (数字)9781665450904

ISBN: (纸本)9781665450904

The area of oil palm plantations in Indonesia increased by 7% from 14 million ha in 2017 to 15 million ha in 2021. The vast land requires the support of effective and efficient management techniques to maintain sustainable productivity. The high-performance computing technologies, Internet of Things (IoT), Big Data, Artificial Intelligence, spatial modeling, and drones are the answers to these needs. This study aims to design an application that can obtain plantation image data and analyze the calculation of the number of oil palm trees. The image of oil palm plantations is obtained from processing photo data from drones through a mosaic and composite process. This study also employed Scrum and UML as a system model development method. The image of the oil palm plantation area is used to build a tree counting model using the Viola-Jones algorithm. The oil palm tree count information generated by this system can then be used by management for fertilizing, harvesting, and monitoring the condition of oil palm trees. © 2022 IEEE.

关键词： Drones

FBG-Based Online Learning and 3-D Shape Control of Unmodeled Continuum and Soft Robots in Unstructured Environments

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lu, Yiang Chen, Wei Lu, Bo Zhou, Jianshu Chen, Zhi Dou, Qi Liu, Yun-Hui T Stone Robotics Institute Department of Mechanical and Automation Engineering The Chinese University of Hong Kong Hong Kong The Department of Computer Science and Engineering The Chinese University of Hong Kong Hong Kong The Hong Kong Center for Logistics Robotics Hong Kong The Robotics and Microsystems Center School of Mechanical and Electric Engineering Soochow University Jiangsu Suzhou China

In this paper, we present a novel and generic data-driven method to servo-control the 3-D shape of continuum and soft robots embedded with fiber Bragg grating (FBG) sensors. Developments of 3-D shape perception and control technologies are crucial for continuum robots to perform the tasks autonomously in surgical interventions. However, owing to the nonlinear properties of continuum robots, one main difficulty lies in the modeling of them, especially for soft robots with variable stiffness. To address this problem, we propose a versatile learning-based adaptive controller by leveraging FBG shape feedback that can online estimate the unknown model of continuum robot against unexpected disturbances and exhibit an adaptive behavior to the unmodeled system without priori data exploration. Based on a new composite adaptation algorithm, the asymptotic convergences of the closed-loop system with learning parameters have been proven by Lyapunov theory. To validate the proposed method, we present a comprehensive experimental study by using two continuum robots both integrated with multicore FBGs, including a robotic-assisted colonoscope and multi-section extensible soft manipulators. The results demonstrate the feasibility, adaptability, and superiority of our controller in various unstructured environments as well as phantom experiments. Copyright © 2022, The Authors. All rights reserved.

关键词： Fiber Bragg gratings