检索结果-内蒙古大学图书馆

Asymmetric Transfer Hashing with Adaptive Bipartite Graph Learning

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lu, Jianglin Zhou, Jie Chen, Yudong Pedrycz, Witold Hung, Kwok-Wai The College of Computer Science and Software Engineering Shenzhen University Guangdong Shenzhen518060 China Department of Electrical and Computer Engineering Northeastern University BostonMA02115 United States The College of Computer Science and Software Engineering Shenzhen University SZU Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society Guangdong Shenzhen518060 China The School of Information Technology and Electrical Engineering University of Queensland BrisbaneQLD4072 Australia The Department of Electrical & Computer Engineering University of Alberta Edmonton Canada The Systems Research Institute Polish Academy of Sciences Warsaw Poland Tencent Music Entertainment Guangdong Shenzhen518060 China

Thanks to the efficient retrieval speed and low storage consumption, learning to hash has been widely used in visual retrieval tasks. However, the known hashing methods assume that the query and retrieval samples lie in homogeneous feature space within the same domain. As a result, they cannot be directly applied to heterogeneous cross-domain retrieval. In this paper, we propose a Generalized Image Transfer Retrieval (GITR) problem, which encounters two crucial bottlenecks: 1) the query and retrieval samples may come from different domains, leading to an inevitable domain distribution gap;2) the features of the two domains may be heterogeneous or misaligned, bringing up an additional feature gap. To address the GITR problem, we propose an Asymmetric Transfer Hashing (ATH) framework with its unsupervised/semi-supervised/supervised realizations. Specifically, ATH characterizes the domain distribution gap by the discrepancy between two asymmetric hash functions, and minimizes the feature gap with the help of a novel adaptive bipartite graph constructed on cross-domain data. By jointly optimizing asymmetric hash functions and the bipartite graph, not only can knowledge transfer be achieved but information loss caused by feature alignment can also be avoided. Meanwhile, to alleviate negative transfer, the intrinsic geometrical structure of single-domain data is preserved by involving a domain affinity graph. Extensive experiments on both single-domain and cross-domain benchmarks under different GITR subtasks indicate the superiority of our ATH method in comparison with the state-of-the-art hashing methods. Copyright © 2022, The Authors. All rights reserved.

关键词： Hash functions

Sensitivity function of LTI fractional order dynamic systems with respect to the orders

学校读者我要写书评

暂无评论

Sensitivity function of LTI fractional order dynamic systems...

American Control Conference (ACC)

作者： Yan Li Dingyü Dingyü Xue YangQuan Chen Center of Self-Organizing & Intelligent Systems CSOIS Electrical an Computer Engineering Department Utah State University Logan UT USA School of Control Science and Engineering Shandong University Jinan Shandong China Institute of Robotics and Artificial Intelligent School of Information Science and Engineering Northeastern University Shenyang China

Many real natural or man-made dynamic systems can be better characterized using a fractional order dynamic model. Since in such a case the order assumes a non-integer value, it is of interest to consider the effect of small perturbation around the nominal value. It is of practical importance to analyze the influence of the order variations on the system behavior. This paper establishes an analytical method for the analysis of the sensitivity function of LTI fractional order dynamic systems with respect to the orders. The continuous dependent condition of the orders of fractional Green's function is derived, which is proved as a necessary and sufficient condition. The singularity of fractional Green's functions is discussed. Several examples are included for illustration.

关键词： Green's function methods Lyapunov method Intelligent robots Sufficient conditions Artificial intelligence Transfer functions Fractional calculus Integral equations Partial differential equations H infinity control

MovieChat+: Question-aware Sparse Memory for Long Video Question Answering

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Song, Enxin Chai, Wenhao Ye, Tian Hwang, Jenq-Neng Li, Xi Wang, Gaoang Zhejiang University-University of Illinois at Urbana-Champaign Institute Zhejiang University China The Department of Electrical and Computer Engineering The University of Washington United States Robotics and Autonomous Systems Thrust Hong Kong University of Science and Technology Guangzhou China The College of Computer Science and Technology Zhejiang University China

Recently, integrating video foundation models and large language models to build a video understanding system can overcome the limitations of specific pre-defined vision tasks. Yet, existing methods either employ complex spatial-temporal modules or rely heavily on additional perception models to extract temporal features for video understanding, and they only perform well on short videos. For long videos, the computational complexity and memory costs associated with long-term temporal connections are significantly increased, posing additional challenges. Taking advantage of the Atkinson-Shiffrin memory model, with tokens in Transformers being employed as the carriers of memory in combination with our specially designed memory mechanism, we propose MovieChat to overcome these challenges. We lift pre-trained multi-modal large language models for understanding long videos without incorporating additional trainable temporal modules, employing a zero-shot approach. Additionally, in our new version, MovieChat+, we design an enhanced vision-question matching-based memory consolidation mechanism to more significantly anchor the predictions of the visual language models in the relevant visual content. MovieChat achieves state-of-the-art performance in long video understanding, along with the released MovieChat-1K benchmark with 1K long video, 2K temporal grounding labels, and 14K manual annotations for validation of the effectiveness of our method. The code along with the dataset can be accessed via the following link. © 2024, CC BY.

关键词： Zero-shot learning

Modular Adaptive Aerial Manipulation under Unknown Dynamic Coupling Forces

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Yadav, Rishabh Dev Dantu, Swati Pan, Wei Sun, Sihao Roy, Spandan Baldi, Simone Department of Computer Science The Univerity of Manchester United Kingdom Multi-robot systems group Faculty of Electrical Engineering Czech Technical University Prague Czech Republic Department of Cognitive Robotics Delft Univesrity of Technology Netherlands Robotics Research Center International Institute of Information Technology Hyderabad India School of Mathematics Southeast University Nanjing China

Successful aerial manipulation largely depends on how effectively a controller can tackle the coupling dynamic forces between the aerial vehicle and the manipulator. However, this control problem has remained largely unsolved as the existing control approaches either require precise knowledge of the aerial vehicle/manipulator inertial couplings, or neglect the state-dependent uncertainties especially arising during the interaction phase. This work proposes an adaptive control solution to overcome this long standing control challenge without any a priori knowledge of the coupling dynamic terms. Additionally, in contrast to the existing adaptive control solutions, the proposed control framework is modular, that is, it allows independent tuning of the adaptive gains for the vehicle position sub-dynamics, the vehicle attitude sub-dynamics, and the manipulator sub-dynamics. Stability of the closed loop under the proposed scheme is derived analytically, and real-time experiments validate the effectiveness of the proposed scheme over the state-of-the-art approaches. © 2024, CC BY.

关键词： Adaptive control systems

Demonstrating Autonomous 3D Path Planning on a Novel Scalable UGV-UAV Morphing Robot

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Sihite, Eric Slezak, Filip Mandralis, Ioannis Salagame, Adarsh Ramezani, Milad Kalantari, Arash Ramezani, Alireza Gharib, Morteza The Department of Aerospace Engineering California Institute of Technology Pasadena United States The Silicon Synapse Labs Department of Electrical and Computer Engineering Northeastern University Boston United States The Robotics and Autonomous Systems DATA61 Csiro BrisbaneQLD4069 Australia The Jet Propulsion Laboratory Pasadena United States

Some animals exhibit multi-modal locomotion capability to traverse a wide range of terrains and environments, such as amphibians that can swim and walk or birds that can fly and walk. This capability is extremely beneficial for expanding the animal's habitat range and they can choose the most energy efficient mode of locomotion in a given environment. The robotic biomimicry of this multi-modal locomotion capability can be very challenging but offer the same advantages. However, the expanded range of locomotion also increases the complexity of performing localization and path planning. In this work, we present our morphing multi-modal robot, which is capable of ground and aerial locomotion, and the implementation of readily available SLAM and path planning solutions to navigate a complex indoor environment. Copyright © 2023, The Authors. All rights reserved.

关键词： Motion planning

Creating Own AI Datasets from Different Language Sources Efficiently: Advantages, Disadvantages and Best Pracitces to Make Own Datasets 6

学校读者我要写书评

暂无评论

Creating Own AI Datasets from Different Language Sources Eff...

6th IEEE International Conference and Workshop Obuda on electrical and Power engineering, CANDO-EPE 2023

作者： Molnar, Gyorgy Cserko, Jazsef Nagy, Eniko Balogh, Zoltan Óbuda University Kandó Kálmán Faculty of Electrical Engineering Budapest Hungary Széchenyi University János Csere Apáczai Faculty Győr Hungary John von Neumann University Gamf Faculty of Engineering and Computer Science Kecskemét Hungary Győr Hungary Institute of Cyberphysical Systems Óbuda University Antal Bejczy Center for Intelligent Robotics John von Neumann Faculty Budapest Hungary Constantine the Philosopher University in Nitra Faculty of Natural Sciences and Informatics Department of Informatics Nitra94901 Slovakia

ISBN: (纸本)9798350328752

Artificial Intelligence (AI) has become a ubiquitous technology that has the potential to revolutionize many industries. However, AI requires access to vast amounts of data to achieve its full potential. By understanding these data sources and the best practices for working with them, we can unlock the full potential of AI and create groundbreaking solutions that benefit society as a whole. Data sources can be utilized to train AI algorithms, including structured and unstructured data, but finding relevant datasets is challenging. The main problem is the international sources. Most of the datasets are available only in English or other common languages. However these are good sources, but translating them is expensive and time-consuming. This paper shows how to generate own datasets from open-source data with translation. © 2023 IEEE.

关键词： Artificial intelligence

Common spatial pattern with feature scaling (FSc-CSP) for motor imagery classification 17th

学校读者我要写书评

暂无评论

17th International Conference on Asia Simulation, AsiaSim 2017

作者： Prathama, Yohanes de Britto Hertyasta Shapiai, Mohd Ibrahim Aris, Siti Armiza Mohd Ibrahim, Zuwairie Jaafar, Jafreezal Fauzi, Hilman Department of Electronic Systems Engineering Malaysia-Japan International Institute of Technology Universiti Teknologi Malaysia Jalan Sultan Yahya Petra Kuala LumpurWP Kuala Lumpur54100 Malaysia Centre for Artificial Intelligence and Robotics Universiti Teknologi Malaysia Jalan Sultan Yahya Petra Kuala LumpurWP Kuala Lumpur54100 Malaysia Razak School of Engineering and Advanced Technology Universiti Teknologi Malaysia Jalan Sultan Yahya Petra Kuala LumpurWP Kuala Lumpur54100 Malaysia Faculty of Electrical and Electronic Engineering Universiti Malaysia Pahang PekanPahang26600 Malaysia Department of Computer and Information Sciences Faculty of Science and Information Technology Universiti Teknologi Petronas Seri Iskandar Malaysia Faculty of Electrical Engineering Telkom University Bandung Indonesia

ISBN: (纸本)9789811064623

Brain-computer Interface (BCI) is a way to translate human thoughts into computer commands. One of the most popular BCI type is Electroencephalography (EEG)-based BCI, where motor imagery is considered one of the most effective ways. Previously, to extract useful information, various filters are introduced, such as spatial, temporal, and spectral filtering. A spatial filtering algorithm called Common Spatial Pattern (CSP) was developed and known to have excellent performance, especially in motor imagery for BCI application. In general, there are several approaches in improving CSP such as regularization approach, analytic approach, and frequency band selection. In general, the existing techniques for band selection is either to select or reject the band by ignoring the importance of the band. For example, Binary Particle Search Optimization Common Spatial Pattern (BPSO-CSP) was proposed to choose multiple possible best bands to be used in processing the data. In this paper, we propose an algorithm called Feature Scaling Common Spatial Pattern (FSc-CSP) to overcome the problem of feature selection. Instead of selecting features, the proposed algorithm employs a feature scaling system to scale the importance of each band by using Genetic Algorithm (GA) altogether with Extreme Learning Machine (ELM) as classifier, with 1 signifying the most important bands, declining until 0 for the unused bands, as opposed to the 1 and 0 selection system used in BPSO-CSP. Conducted experiments show that by employing feature scaling, better results can be achieved especially compared to vanilla CSP and feature selection with 100 hidden nodes in three from five BCI Competition III datasets IVa, namely aa, aw and ay, with around 5–8% better results compared to vanilla CSP and feature selection. © Springer Nature Singapore Pte Ltd. 2017.

关键词： Brain computer interface

Chitosan for Biofunctionalization of Microsystems

学校读者我要写书评

暂无评论

Chitosan for Biofunctionalization of Microsystems

IEEE/NLM Life Science systems and Applications Workshop

作者： Stephan T. Koev Michael A. Powers Vlad Badilita Hyunmin Yi William E. Bentley Gregory F. Payne Gary W. Rubloff Reza Ghodssi Institute for Systems Research ISR University of Maryland College Park MD USA University System of Maryland Adelphi MD US Department of Materials Science and Engineering University of Maryland College Park MD USA Department of Chemical Engineering University of Maryland College Park MD 20742 University of Maryland Biotechnology Institute (UMBI) University of Maryland College Park MD 20742 Bioengineering Graduate Program University of Maryland College Park MD 20742 University of Maryland College Park MD USA University of Maryland Biotechnology Institute College Park MD USA Institute for Systems Research (ISR) University of Maryland College Park MD 20742 Department of Materials Science and Engineering University of Maryland College Park MD 20742 Department of Electrical and Computer Engineering University of Maryland College Park MD 20742 Institute for Systems Research (ISR) University of Maryland College Park MD 20742

ISBN: (纸本)1424402778

We present the use of the polysaccharide chitosan for immobilizing biomolecules on microfabricated device surfaces. The main advantages of chitosan are its abundance of primary amine groups and its ability to be electrodeposited. Biomolecules are easily attached to chitosan 's amines by standard glutaraldehyde chemistry. The electrodeposition of chitosan allows accurate spatial and temporal control of biomolecule placement. We have demonstrated this biofunctionalization method for two different bioMEMS devices: a biophotonic sensor and a micromechanical biosensor. We have successfully assembled probe DNA on each of them and detected hybridization with target DNA as a demonstration of sensor operation. Here, we briefly present the fabrication procedure and testing results for both of these bioMEMS sensing devices

关键词： DNA Biosensors Molecular biophysics Probes Micromechanical devices Optical sensors Optical surface waves Optical interferometry Assembly Biomedical optical imaging

Time dilation visualization in relativity

学校读者我要写书评

暂无评论

Time dilation visualization in relativity

Supercomputing Conference

作者： P.-K. Hsiung R.H. Thibadeau C.B. Cox R.H.P. Dunn Department of Electrical & Computer Engineering Carnegie Mellon University Pittsburgh PA USA Imaging Systems Laboratory The Robotics Institute Carnegie Mellon University Pittsburgh PA USA Department of Physics Carnegie Mellon University Pittsburgh PA USA Department of Art Carnegie Mellon University Pittsburgh PA USA

The authors extend their previous effort in visualizing the spatial aspect of relativistic effects, and treat the phenomenon of time dilation, an inherent temporal effect of special relativity. Here, the authors demonstrate through still-frame images and live animations that, in observing the viewing-independent time dilation, the finite light transit time involved in performing the observation makes the observed time dilation also depend on the viewing condition. As the authors introduce the physics of special relativity into ray-tracing and make time to pass as a ray travels through space, they are able to link the spatial and temporal dimensions in a fundamental and consistent way in their simulations, and generate images that reveal the spatial and temporal properties of the 4D geometry of spacetime. The present exploration highlights the interplay of the imaging process and the imaged physical events in producing the final images.< >

关键词： Visualization Physics Time measurement Ray tracing Space exploration Geometry Clocks Laboratories Orbital robotics Motion measurement