检索结果-南通市图书馆

Quantum-enhanced reinforcement learning for control:a preliminary study

在线全文

学校读者我要写书评

暂无评论

Control Theory and Technology 2021年第4期19卷 455-464页

作者： Yazhou Hu fengzhen tang Jun Chen Wenxue Wang College of Mechanical and Electronic Engineering Northwest A&F UniversityYanglingShaanxi 712100China The State Key Laboratory of Robotics Shenyang Institute of AutomationChinese Academy of SciencesShenyangLiaoning 110016China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of SciencesShenyangLiaoning 110169China

Reinforcement learning is one of the fastest growing areas in machine learning,and has obtained great achievements in biomedicine,Internet of Things(IoT),logistics,robotic control,***,there are still many challenges for engineering applications,such as how to speed up the learning process,how to balance the trade-of between exploration and *** technology,which can solve complex problems faster than classical methods,especially in supercomputers,provides us a new paradigm to overcome these challenges in reinforcement *** this paper,a quantum-enhanced reinforcement learning is pictured for optimal *** this algorithm,the states and actions of reinforcement learning are quantized by quantum *** then,a probability amplifcation method,which can efectively avoid the trade-of between exploration and exploitation via quantized technology,is ***,the optimal control policy is learnt during the process of reinforcement *** performance of this quantized algorithm is demonstrated in both MountainCar reinforcement learning environment and CartPole reinforcement learning environment—one kind of classical control reinforcement learning environment in the OpenAI *** preliminary study results validate that,compared with Q-learning,this quantized reinforcement learning method has better control performance without considering the trade-of between exploration and *** learning performance of this new algorithm is stable with diferent learning rates from 0.01 to 0.10,which means it is promising to be employed in unknown dynamics systems.

关键词： Quantum theory Reinforcement learning Quantum computation State superposition Optimal control

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

面向RatSLAM的自主探索算法

沈阳理工大学学报 2024年第6期43卷 1-7页

作者：武皓武彤张志慧唐凤珍沈阳理工大学自动化与电气工程学院沈阳110159 中国科学院沈阳自动化研究所机器人学国家重点实验室沈阳110016

类脑同步定位与建图(simultaneous localization and mapping,SLAM)通过模仿大脑神经导航机制实现环境建模,同时获得环境中的位置信息,具有较好的导航能力与较高的鲁棒性,但当前的类脑SLAM系统大都在预先获取的环境数据集上进行定位建图... 详细信息

类脑同步定位与建图(simultaneous localization and mapping,SLAM)通过模仿大脑神经导航机制实现环境建模,同时获得环境中的位置信息,具有较好的导航能力与较高的鲁棒性,但当前的类脑SLAM系统大都在预先获取的环境数据集上进行定位建图,对于新环境的兼容性较差。为此提出一种主动的类脑SLAM方法,设计基于Bug算法的环境自主探索方法,驱动机器人在环境中自主移动,主动获取环境信息,同时调用经典的类脑SLAM(RatSLAM)方法对环境信息进行处理,形成环境的认知地图。仿真环境和真实环境下的实验结果表明,本文方法取得了较好的建图效果,与通过人工采集数据建立的环境地图相似度较高,具备可行性.

关键词：自主探索 RatSLAM Bug算法路径规划

维普期刊数据库博看期刊

Principle of Hydrogen Isotope Geochemistry Paleo-altimeter and its Potential in Reconstructing Paleo-elevation of the Southeastern Tibetan Plateau

在线全文

学校读者我要写书评

暂无评论

Acta Geologica Sinica(English Edition) 2024年第4期98卷 1051-1063页

作者： CUI fengzhen LIU-ZENG Jing LI Yunshuai XU Qiang tang Maoyun WANG Heng SUN Zhaotong School of Earth System Science Tianjin UniversityTianjin 300072China State Key Laboratory of Earthquake Dynamics Institute of GeologyChina Earthquake AdministrationBeijing 100029China School of Earth Science and Technology Southwest Petroleum UniversityChengdu 610500China Chongqing Earthquake Agency Chongqing 401147China

The reconstruction of paleo-elevation serves a dual purpose to enhance our comprehension of geodynamic processes affecting terrestrial landforms and to contribute significantly to the interpretation of atmospheric circulation and *** oxygen(δ~(18)O_w)and deuterium(δD_w)isotopes in atmospheric precipitation are systematically depleted with the increase of altitude,which are typical and widely applicated *** utilization of hydrogen isotope of hydrous silicate minerals within the shear zone system,volcanic glass,and plant leaf wax alkanes offers valuable insights for addressing evaporation and *** this paper,we review the principle,application conditions,and influencing factors of the hydrogen isotope *** addition,we discuss the feasibility of utilizing this technique for quantitatively estimating the paleo-elevation of the southeastern Tibetan Plateau,where multiple shear zones extend over hundred kilometers parallel to the topographic gradient.

关键词： paleo-elevation hydrogen isotope hydrous silicate minerals volcanic glass Tibetan Plateau

在线全文

学校读者我要写书评

暂无评论

基于自适应编码的脉冲神经网络

计算机应用研究 2022年第2期39卷 593-597页

作者：张驰唐凤珍中国科学院沈阳自动化研究所机器人学国家重点实验室沈阳110169 中国科学院机器人与智能制造创新研究院沈阳110169 中国科学院大学北京100049

脉冲神经网络(SNN)采用脉冲序列表征和传递信息,与传统人工神经网络相比更具有生物可解释性,但典型SNN的特征提取能力受到其结构限制,对于图像数据等多分类任务的识别准确率不高,不能与卷积神经网络相媲美。为此提出一种新型的自适应编... 详细信息

脉冲神经网络(SNN)采用脉冲序列表征和传递信息,与传统人工神经网络相比更具有生物可解释性,但典型SNN的特征提取能力受到其结构限制,对于图像数据等多分类任务的识别准确率不高,不能与卷积神经网络相媲美。为此提出一种新型的自适应编码脉冲神经网络(SCSNN),将CNN的特征提取能力与SNN的生物可解释性结合起来,采用生物神经元动态脉冲触发特性构建网络结构,并设计了一种新的替代梯度反向传播方法直接训练网络参数。所提出的SCSNN分别在MNIST和Fashion-MNIST数据集进行验证,取得较好的识别结果,在MNIST数据集上准确率达到了99.62%,在Fashion-MNIST数据集上准确率达到了93.52%,验证了其有效性。

关键词：脉冲神经网络自适应编码替代梯度反向传播漏电积分发放神经元模型

在线全文

学校读者我要写书评

暂无评论

基于快速增量式视觉感知的类脑SLAM

信息与控制 2022年第5期51卷 542-553页

作者：徐剑君商亮唐凤珍中国科学院沈阳自动化研究所机器人学国家重点研究室辽宁沈阳110016 中国科学院机器人与智能制造创新研究院辽宁沈阳110169 中国科学院大学北京100049

传统的RatSLAM算法中视觉处理受环境、光照的影响大,进而导致建图精度及稳定性下降。因此,提出了一种快速增量式视觉处理方法克服原RatSLAM系统中的视觉处理的缺陷。以一个改进型的二叉搜索树为检索算法,通过动态岛屿机制对图像进行分组... 详细信息

传统的RatSLAM算法中视觉处理受环境、光照的影响大,进而导致建图精度及稳定性下降。因此,提出了一种快速增量式视觉处理方法克服原RatSLAM系统中的视觉处理的缺陷。以一个改进型的二叉搜索树为检索算法,通过动态岛屿机制对图像进行分组,最终通过序列匹配的形式实现环境识别,达到了在线、准确、快速识别环境的目的。实验结果表明,所提算法的位置识别准确率高于99%,召回率高于80%,平均处理时间低于50 ms。本系统的闭环性能、时间性能及建图稳定性均显著优于现有方案,进一步证明了基于快速增量式视觉处理方法的鲁棒性、高效性。

关键词： RatSLAM 闭环检测即时定位与地图构建

基于对数欧氏度量学习的概率黎曼空间量化方法

在线全文

学校读者我要写书评

暂无评论

计算机应用研究 2022年第3期39卷 661-667,680页

作者：张晓铖唐凤珍中国科学院沈阳自动化研究所机器人学国家重点实验室沈阳110016 中国科学院机器人与智能制造创新研究院沈阳110169 中国科学院大学北京100049

在许多机器学习应用中,需要分析的数据可能由对称正定矩阵构成,而经典的欧氏机器学习算法处理这种数据的性能较差。针对此问题,提出一种新的基于对数欧氏度量学习的概率黎曼空间量化方法。该方法将对称正定矩阵看做对数欧氏度量下黎曼... 详细信息

在许多机器学习应用中,需要分析的数据可能由对称正定矩阵构成,而经典的欧氏机器学习算法处理这种数据的性能较差。针对此问题,提出一种新的基于对数欧氏度量学习的概率黎曼空间量化方法。该方法将对称正定矩阵看做对数欧氏度量下黎曼流形上的点,采用对数欧氏度量学习距离函数将概率学习矢量量化方法从欧氏空间推广到对称正定黎曼空间。在BCI IV 2a脑电数据集上,该方法相较于概率学习矢量量化方法识别正确率提升20%,高于竞赛第一名;并且计算速度快,模型训练及测试时间分别为基于仿射不变度量的同类型算法的1%和10%。在BCI III IIIa和图像数据集ETH-80上也取得了较好的结果。

关键词：对称正定矩阵学习矢量量化对数欧氏度量黎曼几何流形结构

Harboring CAR-T cells using an injectable scaffold to treat solid tumors

在线全文

学校读者我要写书评

暂无评论

National Science Review 2024年第5期11卷 91-92页

作者： Li tang Institute of Bioengineering (IBI) école polytechnique fédérale de Lausanne (EPFL)

Genetic expression of chimeric antigen receptors (CAR) enables T cells to target and kill cancer cells in an MHCindependent manner, resulting in a potent antitumor response [1]. To date, more than 10 CAR-T cell products have been approved globally for the treatment of hematologic disorders such as relapsed/refractory B-cell acute lymphoblastic leukemia, non-Hodgkin lymphoma, and multiple myeloma [2,3].

关键词： IMMUNOTHERAPY CANCER

Topologically protected entanglement switching around exceptional points

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Light(Science & Applications) 2024年第8期13卷 1631-1644页

作者： Zan tang Tian Chen Xing tang Xiangdong Zhang Key Laboratory of Advanced Optoelectronic Quantum Architecture and Measurements of Ministry of Education Beijing Key Laboratory of Nanophotonics&Ultrafine Optoelectronic SystemsSchool of PhysicsBeijing Institute of Technology100081 BeijingChina

The robust operation of quantum entanglement states is crucial for applications in quantum information,computing,and communications^(1–3).However,it has always been a great challenge to complete such a task because of decoherence and ***,we propose theoretically and demonstrate experimentally an effective scheme to realize robust operation of quantum entanglement states by designing quadruple degeneracy exceptional *** encircling the exceptional points on two overlapping Riemann energy surfaces,we have realized a chiral switch for entangled states with high *** to the topological protection conferred by the Riemann surface structure,this switching of chirality exhibits strong robustness against perturbations in the encircling ***,we have experimentally validated such a scheme on a quantum walk *** work opens up a new way for the application of non-Hermitian physics in the field of quantum information.

关键词： exceptional quantum Riemann

An economic way to achieve all-weather CO2 reduction

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

National Science Review 2024年第3期11卷 46-47页

作者： Junwang tang Department of Chemical Engineering University College London

Utilising solar power to convert carbon dioxide（CO2） into hydrocarbon fuels presents an eco-friendly method to address dependence on fossil fuels and contribute to a net-zero CO2emission goal[1,2].

关键词：