检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Zhu, Minjie Zhu, Yichen Li, Jinming Wen, Junjie Xu, Zhiyuan Liu, Ning Cheng, Ran Shen, Chaomin Peng, Yaxin Feng, Feifei Tang, Jian School of Computer Science East China Normal University China Midea Group AI Research Center China Beijing Innovation Center of Humanoid Robotics China Department of Computer Science Shanghai University China

Diffusion Policy is a powerful technique tool for learning end-to-end visuomotor robot control. It is expected that Diffusion Policy possesses scalability, a key attribute for deep neural networks, typically suggesting that increasing model size would lead to enhanced performance. However, our observations indicate that Diffusion Policy in transformer architecture (DP-T) struggles to scale effectively;even minor additions of layers can deteriorate training outcomes. To address this issue, we introduce Scalable Diffusion Transformer Policy for visuomotor learning. Our proposed method, namely ScaleDP, introduces two modules that improve the training dynamic of Diffusion Policy and allow the network to better handle multimodal action distribution. First, we identify that DPT suffers from large gradient issues, making the optimization of Diffusion Policy unstable. To resolve this issue, we factorize the feature embedding of observation into multiple affine layers, and integrate it into the transformer blocks. Additionally, our utilize non-causal attention which allows the policy network to "see" future actions during prediction, helping to reduce compounding errors. We demonstrate that our proposed method successfully scales the Diffusion Policy from 10 million to 1 billion parameters. This new model, named ScaleDP, can effectively scale up the model size with improved performance and generalization. We benchmark ScaleDP across 50 different tasks from MetaWorld and find that our largest ScaleDP outperforms DP-T with an average improvement of 21.6%. Across 7 real-world robot tasks, our ScaleDP demonstrates an average improvement of 36.25% over DP-T on four single-arm tasks and 75% on three bimanual tasks. We believe our work paves the way for scaling up models for visuomotor learning. The project page is available at https://***/. © 2024, CC BY.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Attention in Attention for PET-CT Modality Consensus Lung Tumor Segmentation

Attention in Attention for PET-CT Modality Consensus Lung Tu...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Yuzhou Zhao Xinyu Zhou Haijing Guo Qianyu Guo Yan Zuo Shaoli Song Shuyong Gao Wenqiang Zhang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Department of Nuclear Medicine Fudan University Shanghai Cancer Center China Shanghai Engineering Research Center of AI Robotics Academy for Engineering and Technology Fudan University

ISBN: (数字)9798350390155

ISBN: (纸本)9798350390162

Combination of multi-modal PET-CT imaging for lung tumor segmentation is significant for clinical treatment. Existing methods have not fully considered the impact of noise in PET-CT on the multi-modal interaction. To address this, we propose a novel Attention in Attention Network (aiANet). aiANet can mutually learn multi-modal characteristics for segmentation through its cross-learning modules. Within the cross-learning module, we introduce two nested-attention blocks, namely Attention in Self-Attention (aiSA) and Attention in Cross-Attention (aiCA), for multi-scale feature enhancement and multi-modal feature interaction. Importantly, since traditional attention weights calculated solely or unilaterally based on PET or CT can be vulnerable to the inevitable noisy information, we embed a novel Attention in Attention (aiA) module into aiSA and aiCA. The aiA module can seek cross-modal consensus for attention weights to alleviate their noise. Experimental results on clinical PET-CT data of lung cancer demonstrate the superiority of our method.

关键词： Image segmentation Computed tomography Aggregates Noise Lung cancer Imaging Noise measurement Tumors

来源：评论

学校读者我要写书评

暂无评论

Revolutionizing Intrusive Neural Interfaces: Vascular Stent-Based Neural Electrodes

Revolutionizing Intrusive Neural Interfaces: Vascular Stent-...

引用

International Winter Workshop on Brain-computer Interface (BCI)

作者： Liang Shen Bo Wen Zihua Chen Minjie Wang Linlin Han Bo Shen Kai Cao Lihua Zhang Jianjun Wu Xiaoyang Kang Shanghai Engineering Research Center of AI & Robotics MOE Frontiers Center for Brain Science State Key Laboratory of Medical Neurobiology Laboratory for Neural Interface and Brain Computer Interface Engineering Research Center of AI & Robotics Ministry of Education Institute of AI & Robotics Institute of Meta-Medical Academy for Engineering & Technology Fudan University Shanghai China Department of Neurology and National Research Center for Aging and Medicine & National Center for Neurological Disorders State Key Laboratory of Medical Neurobiology Huashan Hospital Fudan University Shanghai China Yiwu Research Institute of Fudan University Chengbei Road Yiwu City Zhejiang China

This review delves into the emerging field of endovascular stent electrodes in the realm of brain-computer interfaces, highlighting advancements in their fabrication methods, signal acquisition performance, electrical stimulation effects, and research progress in electrochemical characteristics, mechanical stability, and biocompatibility. In comparison to traditional invasive procedures, endovascular stent electrodes, implanted through minimally invasive means, mitigate the risks associated with extensive trauma and tissue inflammation, demonstrating commendable signal acquisition capabilities. While exhibiting a slight deficiency in visual evoked potentials, they showcase accuracy in the acquisition and decoding of motor perception signals, rivaling or surpassing traditional surgical methods. Additionally, the technology shows potential therapeutic efficacy in electrical stimulation applications. Nevertheless, challenges such as electrical stimulation thresholds and visibility issues warrant further research and refinement. In summary, endovascular stent electrodes, as an emerging avenue in brain-computer interfaces, hold promise for significant breakthroughs in future research, opening up new possibilities in the field of neuroscience.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Delving into Decision-based Black-box Attacks on Semantic Segmentation

arXiv

引用

arXiv 2024年

作者： Chen, Zhaoyu Shan, Zhengyang Chang, Jingwen Jiang, Kaixun Yang, Dingkang Cheng, Yiting Zhang, Wenqiang Shanghai Engineering Research Center of AI & Robotics Academy for Engineering & Technology Fudan University China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University China

Semantic segmentation is a fundamental visual task that finds extensive deployment in applications with security-sensitive considerations. Nonetheless, recent work illustrates the adversarial vulnerability of semantic segmentation models to white-box attacks. However, its adversarial robustness against black-box attacks has not been fully explored. In this paper, we present the first exploration of black-box decision-based attacks on semantic segmentation. First, we analyze the challenges that semantic segmentation brings to decision-based attacks through the case study. Then, to address these challenges, we first propose a decision-based attack on semantic segmentation, called Discrete Linear Attack (DLA). Based on random search and proxy index, we utilize the discrete linear noises for perturbation exploration and calibration to achieve efficient attack efficiency. We conduct adversarial robustness evaluation on 5 models from Cityscapes and ADE20K under 8 attacks. DLA shows its formidable power on Cityscapes by dramatically reducing PSPNet’s mIoU from an impressive 77.83% to a mere 2.14% with just 50 queries. Copyright © 2024, The Authors. All rights reserved.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Fast Walking of Position-Controlled Biped Robot Based on Whole-Body Compliance Control 16th

Fast Walking of Position-Controlled Biped Robot Based on Who...

引用

16th International Conference on Intelligent robotics and Applications, ICIRA 2023

作者： Zhu, Yiting Xie, Ye Gao, Chengzhi Liang, Dingkun Kong, Lingyu Wang, Xin Xie, Anhuan Gu, Jianjun Research Center for Intelligent Robotics Research Institute of Interdisciplinary Innovation Zhejiang Laboratory Hangzhou311100 China Zhejiang Engineering Research Center for Intelligent Robotics Hangzhou311100 China Zhejiang University Zhejiang Hangzhou311100 China Department of Electrical and Computer Engineering Dalhousie University HalifaxNSB3H 4R2 Canada

ISBN: (纸本)9789819964918

When a biped robot walks at a high speed, only relying on the open-loop gait trajectory based on the linear inverted pendulum model (LIPM), the robot cannot maintain stability. Sudden contact with the ground when landing may cause the robot to be unstable. Simultaneously, when the robot walks faster, the horizontal position of the center of mass (CoM) varies greatly during a walk cycle, and the expected CoM trajectory cannot be accurately tracked, resulting in a deviation from the zero-moment point (ZMP). aiming at the above situation that may occur in fast walking, a whole-body compliance controller with emphasis on making robot’s walking more stable and faster is presented in this paper. The whole-body compliance controller consists of three parts: the CoM compliance controller, the vertical compliance controller and the foot compliance controller, which are used to improve the tracking effect of CoM, reduce the force impact of foot, and adjust the posture of the foot respectively. The simulation results show that with the whole-body compliance controller proposed in this paper, the tracking accuracy of the CoM becomes better, the posture of the robot is more stable during walking, and the landing compliance during landing are greatly improved. Finally, the walking speed of the robot can reach 1.8m/s. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Compliance control

来源：评论

学校读者我要写书评

暂无评论

Addressing Biases in Gastric Cancer Diagnosis Through Generative Models and Vision-Based Surface Tactile Sensing

Addressing Biases in Gastric Cancer Diagnosis Through Genera...

引用

Medical Imaging 2025: Image-Guided Procedures, Robotic Interventions, and Modeling

作者： Kapuria, Siddhartha Ikoma, Naruhiko Chinchali, Sandeep Alambeigi, Farshid Walker Department of Mechanical Engineering and The Texas Robotics The University of Texas at Austin AustinTX United States Texas Robotics The University of Texas at Austin AustinTX United States Department of Surgical Oncology Division of Surgery The University of Texas MD Anderson Cancer Center HoustonTX United States Chandra Family Department of Electrical and Computer Engineering and The Texas Robotics The University of Texas at Austin AustinTX United States

ISBN: (纸本)9781510685949

Towards developing an informed, intuitive, and generalized artificial intelligence model for endoscopic diagnosis of Advanced Gastric Cancer (AGC) lesions, in this work, we extend the use of generative models to augment limited and biased training datasets. This approach aims to enhance model performance and generalizability, tackling the challenge of acquiring sizable and well-balanced datasets in the clinical domain. We applied this approach to our unique dataset comprised of textural images of additively-manufactured realistic AGC tumor phantoms collected using a robotic data collection system and our unique Vision-based Tactile Sensor - HySenSe. Our approach was particularly suited to this dataset, since HySenSe's limited sensing area captures only partial images of the much larger AGC lesions, which reduces the amount of information available in a single image. We trained a single Class-Conditioned Latent Diffusion Model (CC-LDM) on an artificially imbalanced dataset of HySenSe images, logging performance using Fréchet Inception Distance (FID), and exploring different hyperparameter configurations. Using the high-quality synthetic images generated using this model, we performed experiments using our previously developed Dilated ResNet model for AGC tumor classification with the following objectives: (i) study the amount of synthetic data required during training, and (ii) compare different data addition strategies during model training using cross-validation. Note that while the models were trained using both real and synthetic textural images, testing was performed solely on real images. Our experiments showed that the trained classifier demonstrates improved generalizability when trained on a mixture of real and synthetic images, but not all methods of mixing the data are equally effective. Overall, the model shows improved performance even under mixed morphological tumor conditions and partial sensor contact. © 2025 SPIE

关键词： Tactile sensors

来源：评论

学校读者我要写书评

暂无评论

DARA: Decomposition-Alignment-Reasoning Autonomous Language Agent for Question Answering over Knowledge Graphs

arXiv

引用

arXiv 2024年

作者： Fang, Haishuo Zhu, Xiaodan Gurevych, Iryna Department of Computer Science and Hessian Center for AI hessian.AI Technical University of Darmstadt Germany Department of Electrical and Computer Engineering Ingenuity Labs Research Institute Queen’s University Canada

Answering Questions over Knowledge Graphs (KGQA) is key to well-functioning autonomous language agents in various real-life applications. To improve the neural-symbolic reasoning capabilities of language agents powered by Large Language Models (LLMs) in KGQA, we propose the Decomposition-Alignment-Reasoning Agent (DARA) framework. DARA effectively parses questions into formal queries through a dual mechanism: high-level iterative task decomposition and low-level task grounding. Importantly, DARA can be efficiently trained with a small number of high-quality reasoning trajectories. Our experimental results demonstrate that DARA fine-tuned on LLMs (e.g. Llama-2-7B, Mistral) outperforms both in-context learning-based agents with GPT-4 and alternative fine-tuned agents, across different benchmarks in zero-shot evaluation. This makes such models more accessible for real-life applications. We also show that DARA attains performance comparable to state-of-the-art enumerating-and-ranking-based methods for KGQA1 © 2024, CC BY.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Enhancing Relation Extraction by Using Shortest Dependency Paths Between Entities with Pre-trained Language Models 16

Enhancing Relation Extraction by Using Shortest Dependency P...

引用

16th International Conference on INnovations in Intelligent SysTems and Applications, INISTA 2022

作者： Karaevli, Haluk Alper Gungor, Tunga Boǧaziçi University Department of Computer Engineering İstanbul Turkey Turkey R&d Center Huawei Technologies Ai Enablement Department Turkey

ISBN: (数字)9781665498104

ISBN: (纸本)9781665498104

Relation Extraction (RE) is the task of finding the relation between entities in a plain text. As the length of the sentences increases, finding the relation becomes more challenging. The shortest dependency path (SDP) between two entities, obtained by traversing the terms in the dependency tree of a sentence, provides a view focused on the entities by pruning noisy words. In the supervised form of the relation extraction task, Relation Classification, the state-of-the-art methods generally integrate a pre-trained language model (PLM) into their approaches. However, none of them incorporates the shortest dependency paths to the best of our *** paper investigates the effects of using shortest dependency paths with pre-trained language models by taking the R-BERT relation classification model as the baseline and building upon it. Our novel approach enhances the baseline model by adding the sequence representation of the shortest dependency path between entities, collected from PLMs, as an additional embedding. In the experiments, we evaluated the proposed model's performance for each combination of SDPs generated from Stanford, HPSG, and LAL dependency parsers with BERT and XLNet PLMs in two datasets, SemEval-2010 Task 8 and TACRED. We improved the baseline model by absolute 1.41% and 3.60% scores, increasing the rankings of the model from 8th to 7th and from 18th to 7th in SemEval-2010 Task 8 and TACRED, respectively. © 2022 IEEE.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Non-invasive estimation of the powder size distribution from a single speckle image

引用

Light(Science & Applications) 2024年第10期13卷 2221-2230页

作者： Qihang Zhang Ajinkya Pandit Zhiguang Liu Zhen Guo Shashank Muddu Yi Wei Deborah Pereg Neda Nazemifard Charles Papageorgiou Yihui Yang Wenlong Tang Richard D.Braatz Allan S.Myerson George Barbastathis Department of Electrical Engineering and Computer Science Massachusetts Institute of TechnologyCambridgeMA 02139USA Singapore-MIT Alliance for Research and Technology(SMART)Centre Singapore 117543Singapore Department of Chemical Engineering Massachusetts Institute of TechnologyCambridgeMA 02139USA Department of Mechanical Engineering Massachusetts Institute of TechnologyCambridgeMA 02139USA Process Chemistry Development Takeda Pharmaceuticals International Co40 Landsdowne StCambridgeMA 02139USA ShinrAI Center for AI/ML Data Sciences InstitutesTakeda Pharmaceuticals International Co650 E Kendall StCambridgeMA 02142USA Present address:Department of Precision Instruments Tsinghua UniversityBeijing 100084China

Non-invasive characterization of powders may take one of two approaches:imaging and counting individual particles;or relying on scattered light to estimate the particle size distribution(PSD)of the *** former approach runs into practical difficulties,as the system must conform to the working distance and other restrictions of the imaging *** latter approach requires an inverse map from the speckle autocorrelation to the particle *** principle relies on the pupil function determining the basic sidelobe shape,whereas the particle size spread modulates the sidelobe *** recently showed that it is feasible to invert the speckle autocorrelation and obtain the PSD using a neural network,trained efficiently through a physics-informed semi-generative *** this work,we eliminate one of the most time-consuming steps of our previous method by engineering the pupil *** judiciously blocking portions of the pupil,we sacrifice some photons but in return we achieve much enhanced sidelobes and,hence,higher sensitivity to the change of the size *** result is a 60×reduction in total acquisition and processing time,or 0.25 seconds per frame in our *** real-time operation in our system is not only more appealing toward rapid industrial adoption,it also paves the way for quantitative characterization of complex spatial or temporal dynamics in drying,blending,and other chemical and pharmaceutical manufacturing processes.

关键词： speckle size distribution

来源：评论

学校读者我要写书评

暂无评论

Wireless Sensor Networks and Green Networking for 6G communication- An Overview

Wireless Sensor Networks and Green Networking for 6G communi...

引用

Computational Intelligence, Communication Technology and Networking (CICTN), International Conference on

作者： Ramiz Salama Fadi Al-Turjman Dibyahash Bordoloi Satya Prakash Yadav Department of Computer Engineering AI and Robotics Institute Research Center for AI and IoT Near East University Nicosia Turkey Artificial Intelligence Engineering Dept. AI and Robotics Institute Near East University Nicosia Turkey Faculty of Engineering Research Center for AI and IoT University of Kyrenia Kyrenia Turkey Dept. of Computer Science and Engineering Graphic Era Hill University Dehradun RS CSE Graphic Era University Dehradun Department of Computer Science and Engineering G.L. Bajaj Institute of technology and Management (GLBITM) Greater Noida India

Wireless sensor networks and green networking have been major research areas in the field of communication technology for the past few decades. In particular, the development of 6G communication systems has brought renewed focus on these areas, as the demands for higher data rates, more diverse applications, and lower energy consumption continue to increase. The use of wireless sensor networks in 6G communication systems has been widely explored in the literature, and has shown great potential for a wide range of applications, including environmental monitoring, industrial process control, and healthcare. However, a key challenge in the deployment of these networks is ensuring their energy efficiency, as the demands on the network will be much higher than in previous generations. To address this challenge, researchers have proposed various green networking techniques, such as energy harvesting and low-power communication protocols, which have been shown to significantly reduce the energy consumption of the sensors. In addition to improving the energy efficiency of wireless sensor networks, researchers have also studied their reliability, scalability, and adaptability. These properties are important in 6G communication systems, as the demands on the network are expected to be much higher than in previous generations. Researchers have proposed various methods to improve the reliability, scalability, and adaptability of wireless sensor networks, including the use of error-correction codes, redundant communication channels, and machine learning algorithms.

关键词：

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：