检索结果-内蒙古大学图书馆

Integrating Generative Artificial Intelligence Into Future Internet of Vehicles [From the Guest Editors]

学校读者我要写书评

暂无评论

IEEE Vehicular Technology Magazine 2025年第2期20卷 40-42页

作者： Sun, Gang Du, Hongyang Jamalipour, Abbas Muntean, Gabriel-Miro Cai, Lin Kang, Jiawen Niyato, Dusit University of Electronic Science and Technology of China Chengdu611731 China The University of Hong Kong Department of Electrical and Electronic Engineering 999077 Hong Kong University of Sydney SydneyNSW2006 Australia Dublin City University Performance Engineering Laboratory DublinD9 Ireland University of Victoria Department of Electrical and Computer Engineering VictoriaBCV8W 2Y2 Canada Guangdong University of Technology Guangzhou510006 China Nanyang Technological University Singapore College of Computing and Data Science 639798 Singapore

ARTxAI: Explainable Artificial Intelligence Curates Deep Representation Learning for Artistic Images using Fuzzy Techniques

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Fumanal-Idocin, Javier Andreu-Perez, J. Cordón, Oscar Hagras, H. Bustince, Humberto The Departamento de Estadistica Informatica y Matematicas Universidad Pub-lica de Navarra Campus de Arrosadia Pamplona31006 Spain University of Granada Granada18071 Spain The School of Computer Science and Electronic Engineering University of Essex Colchester United Kingdom Simbad2 Department of Computer Science University of Jaen Jaen Spain

Automatic art analysis employs different image processing techniques to classify and categorize works of art. When working with artistic images, we need to take into account further considerations compared to classical image processing. This is because such artistic paintings change drastically depending on the author, the scene depicted, and their artistic style. This can result in features that perform very well in a given task but do not grasp the whole of the visual and symbolic information contained in a painting. In this paper, we show how the features obtained from different tasks in artistic image classification are suitable to solve other ones of similar nature. We present different methods to improve the generalization capabilities and performance of artistic classification systems. Furthermore, we propose an explainable artificial intelligence method to map known visual traits of an image with the features used by the deep learning model considering fuzzy rules. These rules show the patterns and variables that are relevant to solve each task and how effective is each of the patterns found. Our results show that our proposed context-aware features can achieve up to 6% and 26% more accurate results than other context- and non-context-aware solutions, respectively, depending on the specific task. We also show that some of the features used by these models can be more clearly correlated to visual traits in the original image than others. © 2023, CC BY.

关键词： Fuzzy rules

Nonparametric Spatio-Temporal Joint Probabilistic Data Association Coupled Filter and Interfering Extended Target Tracking

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Akbari, Behzad Zhu, Haibin Pan, Ya-Jun Tharmarasa, R. Department of Mechanical Engineering Dalhousie University HalifaxNSB3H 4R2 Canada Department of Computer Science and Mathematics Nipissing University North BayONP1B 8L7 Canada Electrical and Computer Engineering Department McMaster University HamiltonONL8S 4K1 Canada

Extended target tracking estimates the centroid and shape of the target in space and time. In various situations where extended target tracking is applicable, the presence of multiple targets can lead to interference, particularly when they maneuver behind one another in a sensor like a camera. Nonetheless, when dealing with multiple extended targets, there's a tendency for them to share similar shapes within a group, which can enhance their detectability. For instance, the coordinated movement of a cluster of aerial vehicles might cause radar misdetections during their convergence or divergence. Similarly, in the context of a self-driving car, lane markings might split or converge, resulting in inaccurate lane tracking detections. A well-known joint probabilistic data association coupled (JPDAC) filter can address this problem in only a single-point target tracking. A variation of JPDACF was developed by introducing a nonparametric Spatio-Temporal Joint Probabilistic Data Association Coupled Filter (ST-JPDACF) to address the problem for extended targets. Using different kernel functions, we manage the dependency of measurements in space (inside a frame) and time (between frames). Kernel functions are able to be learned using a limited number of training data. This extension can be used for tracking the shape and dynamics of nonparametric dependent extended targets in clutter when targets share measurements. The proposed algorithm was compared with other well-known supervised methods in the interfering case and achieved promising results. Copyright © 2023, The Authors. All rights reserved.

关键词： Clutter (information theory)

Position: explain to question not to justify 24

学校读者我要写书评

暂无评论

Position: explain to question not to justify

Proceedings of the 41st International Conference on Machine Learning

作者： Przemyslaw Biecek Wojciech Samek MI2.AI University of Warsaw Poland and Warsaw University of Technology Poland Department of Artificial Intelligence Fraunhofer Heinrich Hertz Institute Germany and Department of Electrical Engineering and Computer Science Technical University of Berlin Germany and BIFOLD - Berlin Institute for the Foundations of Learning and Data Germany

Explainable Artificial Intelligence (XAI) is a young but very promising field of research. Unfortunately, the progress in this field is currently slowed down by divergent and incompatible goals. We separate various threads tangled within the area of XAI into two complementary cultures of human/value-oriented explanations (BLUE XAI) and model/validation-oriented explanations (RED XAI). This position paper argues that the area of RED XAI is currently under-explored, i.e., more methods for explainability are desperately needed to question models (e.g., extract knowledge from well-performing models as well as spotting and fixing bugs in faulty models), and the area of RED XAI hides great opportunities and potential for important research necessary to ensure the safety of AI systems. We conclude this paper by presenting promising challenges in this area.

关键词：

Solving Maxwell's equations with Non-Trainable Graph Neural Network Message Passing

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Bakirtzis, Stefanos Fiore, Marco Zhang, Jie Wassell, Ian The Department of Computer Science and Technology University of Cambridge CambridgeCB3 0FD United Kingdom The IMDEA Networks Institute Madrid Spain The Department of Electronic and Electrical Engineering University of Sheffield SheffieldS10 2TN United Kingdom

Computational electromagnetics (CEM) is employed to numerically solve Maxwell’s equations, and it has very important and practical applications across a broad range of disciplines, including biomedical engineering, nanophotonics, wireless communications, and electrodynamics. The main limitation of existing CEM methods is that they are computationally demanding. Our work introduces a leap forward in scientific computing and CEM by proposing an original solution of Maxwell’s equations that is grounded on graph neural networks (GNNs) and enables the high-performance numerical resolution of these fundamental mathematical expressions. Specifically, we demonstrate that the update equations derived by discretizing Maxwell’s partial differential equations can be innately expressed as a two-layer GNN with static and pre-determined edge weights. Given this intuition, a straightforward way to numerically solve Maxwell’s equations entails simple message passing between such a GNN’s nodes, yielding a significant computational time gain, while preserving the same accuracy as conventional transient CEM methods. Ultimately, our work supports the efficient and precise emulation of electromagnetic wave propagation with GNNs, and more importantly, we anticipate that applying a similar treatment to systems of partial differential equations arising in other scientific disciplines, e.g., computational fluid dynamics, can benefit computational sciences. © 2024, CC BY-NC-SA.

关键词： Computational electromagnetics

MetaGCD: Learning to Continually Learn in Generalized Category Discovery

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wu, Yanan Chi, Zhixiang Wang, Yang Feng, Songhe Key Laboratory of Big Data & Artificial Intelligence in Transportation Ministry of Education Beijing Jiaotong University Beijing100044 China School of Computer and Information Technology Beijing Jiaotong University Beijing100044 China Department of Electrical and Computer Engineering University of Toronto TorontoM5G1V7 Canada Department of Computer Science and Software Engineering Concordia University MontrealH3G2J1 Canada

In this paper, we consider a real-world scenario where a model that is trained on pre-defined classes continually encounters unlabeled data that contains both known and novel classes. The goal is to continually discover novel classes while maintaining the performance in known classes. We name the setting Continual Generalized Category Discovery (C-GCD). Existing methods for novel class discovery cannot directly handle the C-GCD setting due to some unrealistic assumptions, such as the unlabeled data only containing novel classes. Furthermore, they fail to discover novel classes in a continual fashion. In this work, we lift all these assumptions and propose an approach, called MetaGCD, to learn how to incrementally discover with less forgetting. Our proposed method uses a meta-learning framework and leverages the offline labeled data to simulate the testing incremental learning process. A meta-objective is defined to revolve around two conflicting learning objectives to achieve novel class discovery without forgetting. Furthermore, a soft neighborhood-based contrastive network is proposed to discriminate uncorrelated images while attracting correlated images. We build strong baselines and conduct extensive experiments on three widely used benchmarks to demonstrate the superiority of our method. Our code is available at https://***/ynanwu/MetaGCD. © 2023, CC0.

关键词： Learning systems

Delay Constraint Energy Efficient Cooperative Offloading in MEC for IoT 16th

学校读者我要写书评

暂无评论

Delay Constraint Energy Efficient Cooperative Offloading in ...

16th EAI International Conference on Collaborative Computing: Networking, Applications, and Worksharing, CollaborateCom 2020

作者： Sun, Haifeng Wang, Jun Peng, Haixia Song, Lili Qin, Mingwei School of Computer Science and Technology Southwest University of Science and Technology Mianyang621010 China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518061 China Department of Electrical and Computer Engineering University of Waterloo WaterlooN2L 3G1 Canada School of Information Engineering Southwest University of Science and Technology Mianyang621010 China

ISBN: (纸本)9783030675363

Mobile edge computing (MEC) is a promising approach to execute delay-sensitive and computation-intensive applications in the resource-limited IoT mobile devices (IMDs) by offloading computing tasks to MEC servers. In this paper, we propose a neighbor-aided cooperative offloading scheme with delay constraint to improve the energy efficiency in MEC-based Internet of Things (IoT) networks. The network consists of a target IMD with some IMD neighbors, and an access point integrated with an MEC server. The latency-constrained tasks in the IMD can be partially offloaded to and executed by the selected neighboring IMD or the MEC server. Different from other works, our proposed offloading scheme selects the most energy-efficient one from all the neighboring IMDs as the offloading helper. Specifically, we formulate an optimization problem to minimize the total energy consumption while satisfying the computation delay constraint of each task, and obtain the most energy-efficient neighbor with the optimized division of tasks by solving the formulated problem. Moreover, we design an easy neighbor selection scheme with lower time complexity by the weighted value of the transmission rate for each neighbor. Numerical results show that the proposed scheme outperforms benchmark schemes significantly in terms of energy consumption and the supported maximum task length. © 2021, ICST Institute for computer sciences, Social Informatics and Telecommunications engineering.

关键词： Energy efficiency

Light propagation in time-periodic bi-isotropic media

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Koufidis, Stefanos Fr. Koutserimpas, Theodoros T. Monticone, Francesco McCall, Martin W. Blackett Laboratory Department of Physics Imperial College of Science Technology and Medicine Prince Consort Road LondonSW7 2AZ United Kingdom School of Electrical and Computer Engineering Cornell University IthacaNY14853 United States

Photonic structures and time-crystals, wherein time is incorporated as an additional degree of freedom for light manipulation, have necessitated the development of analytical and semi-analytical tools. However, such tools are currently limited to specific configurations, leaving several unexplored physical phenomena akin to photonic time-crystals elusive. In this communication, using a coupled-wave theory approach, we unveil the occurring light propagation phenomena in a time-periodic bi-isotropic medium whose permittivity, permeability, and chirality parameter are periodic functions of time. Contrary to their static counterparts, we demonstrate that the considered dynamic medium couples only co-handed counter-propagating waves. In cases of non-constant impedance, we prove that two first-order momentum gaps are formed in the Brillouin diagram, resulting in parametric amplification with different amplification factors and corresponding momenta for the right- and left-handed modes, respectively. The presence of chirality plays a major role in manipulating lightwave signals by controlling the center of resonance, the corresponding bandwidth, and the amplification factor in a distinct fashion for each mode. For a finite "time-slab" of the medium, we analytically derive the scattering coefficients as functions of time and momentum, discussing how extreme values of optical rotation grant access to the temporal analog of the chirality-induced negative refraction regime. Finally, we demonstrate the mechanism under which elliptical polarizations may change field orientation whilst the electric field propagates in a momentum gap, thus simultaneously showcasing parametric amplification. Copyright © 2024, The Authors. All rights reserved.

关键词： Degrees of freedom (mechanics)

N-DRIVERMOTION: DRIVER MOTION LEARNING AND PREDICTION USING AN EVENT-BASED CAMERA AND DIRECTLY TRAINED SPIKING NEURAL NETWORKS

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Chung, Hyo Jong Kang, Byungkon Yang, Yoonseok College of Engineering and Applied Sciences Stony Brook University 100 Nicolls Road New York United States Department of Computer Science The State University of New York Korea 119-2 Songdo Munhwa-ro Yeonsu-gu Incheon Korea Republic of

Driver motion recognition is a key factor in ensuring the safety of driving systems. This paper presents a novel system for learning and predicting driver motions, along with an event-based (720x720) dataset, N-DriverMotion, newly collected to train a neuromorphic vision system. The system includes an event-based camera that generates a driver motion dataset representing spike inputs and efficient spiking neural networks (SNNs) that are effective in training and predicting the driver’s gestures. The event dataset consists of 13 driver motion categories classified by direction (front, side), illumination (bright, moderate, dark), and participant. A novel optimized four-layer convolutional spiking neural network (CSNN) was trained directly without any time-consuming preprocessing. This enables efficient adaptation to energy- and resource-constrained on-device SNNs for real-time inference on high-resolution event-based streams. Compared to recent gesture recognition systems adopting neural networks for vision processing, the proposed neuromorphic vision system achieves competitive accuracy of 94.04% in a 13-class classification task, and 97.24% in an unexpected abnormal driver motion classification task with the CSNN architecture. Additionally, when deployed to Intel Loihi 2 neuromorphic chips, the energy-delay product (EDP) of the model achieved 20,721 times more efficient than that of a non-edge GPU, and 541 times more efficient than edge-purpose GPU. Our proposed CSNN and the dataset can be used to develop safer and more efficient driver-monitoring systems for autonomous vehicles or edge devices requiring an efficient neural network *** Codes 68T45 © 2024, CC BY.

关键词： Gesture recognition