检索结果-南通市图书馆

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

国标麻将的多尺度骨干神经网络模型

重庆理工大学学报（自然科学） 2024年第5期38卷 137-144页

作者：代君学李霞丽刘博王昭琦中央民族大学民族语言智能分析与安全治理教育部重点实验室北京100081 中央民族大学信息工程学院北京100081

在有多轮次、状态空间巨大、81种不同类别的番种、胡牌方式复杂的国标麻将中,普通的神经网络难以对复杂的数据进行表达和拟合。首次将多尺度骨干的深度神经网络用于实现麻将AI,以更好地捕获国标麻将的局部以及全局特征,适用于处理复杂数... 详细信息

在有多轮次、状态空间巨大、81种不同类别的番种、胡牌方式复杂的国标麻将中,普通的神经网络难以对复杂的数据进行表达和拟合。首次将多尺度骨干的深度神经网络用于实现麻将AI,以更好地捕获国标麻将的局部以及全局特征,适用于处理复杂数据,做出更准确的游戏策略。基于IJCAI 2020 Champion的对局数据,对训练数据进行数据增强。采用增强后的数据,在NVIDAI GeForce RTX3090 LapTop GPU上进行了5天的监督学习训练,训练出的模型有52 M参数,动作准确率达到93.47%,弃牌准确率达到83.93%,鸣牌准确率达到97.56%。将提出的模型部署到北京大学开发的Botzone平台上,进入天梯榜前1%。

关键词：深度学习麻将卷积神经网络 Res2Net50 多尺度骨干架构

A modified YOLOv4 detection method for a vision-based underwater garbage cleaning robot

维普期刊数据库

同方期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2022年第8期23卷 1217-1228页

作者： Manjun TIAN xiali li Shihan KONG licheng WU Junzhi YU First Research Institute of the Ministry of Public Security of PRC Beijing 100048China School of Information Engineering Minzu University of ChinaBeijing 100081China Department of Advanced Manufacturing and Robotics College of EngineeringPeking UniversityBeijing 100871China State Key Laboratory of Management and Control for Complex Systems Institute of AutomationChinese Academy of SciencesBeijing 100190China

To tackle the problem of aquatic environment pollution,a vision-based autonomous underwater garbage cleaning robot has been developed in our *** propose a garbage detection method based on a modified YOLOv4,allowing high-speed and high-precision object ***,the YOLOv4 algorithm is chosen as a basic neural network framework to perform object *** the purpose of further improvement on the detection accuracy,YOLOv4 is transformed into a four-scale detection *** improve the detection speed,model pruning is applied to the new *** virtue of the improved detection methods,the robot can collect garbage *** detection speed is up to 66.67 frames/s with a mean average precision(mAP)of 95.099%,and experimental results demonstrate that both the detection speed and the accuracy of the improved YOLOv4 are excellent.

关键词： Object detection Aquatic environment Garbage cleaning robot Modified YOLOv4

A new escape time algorithm of constructing Julia set

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

High Technology Letters 2007年第2期13卷 194-197页

作者：袁杰 li xiali Hou Zhiling Cao Maosheng School of Mathematics and Computer Science Central University of Nationalities Beijing 100081 P.R.China School of Materials Science and Engineering Beijing Institute of Technology Beijing 100081 P.R.China

Escape time algorithm is an effective theoretical algorithm of constructing fractal graphics. The key of this algorithm lies in the construction of escape time function. A new escape time function is presented based on the research of escape time algorithm. An accelerated escape time algorithm is carried out in this paper. The experiments have demonstrated that the new algorithm is not only as precise as the old, but also faster when it is used to construct Julia set.

关键词：逸出时间算法函数功能集合

In situ injectable hydrogel encapsulating Mn/NO-based immune nano-activator for prevention of postoperative tumor recurrence

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

Asian Journal of Pharmaceutical Sciences 2024年第2期19卷 102-119页

作者： Shengnan Huang Chenyang Zhou Chengzhi Song xiali Zhu Mingsan Miao Chunming li Shaofeng Duan Yurong Hu Academy of Chinese Medical Sciences Henan University of Chinese MedicineZhengzhou 450046China School of Pharmaceutical Sciences Henan Key Laboratory of Targeting Therapy and Diagnosis for Critical DiseasesZhengzhou UniversityZhengzhou 450001China Center for Quantitative Biology Peking UniversityBeijing 100871China School of Pharmaceutical Sciences Henan University of Chinese MedicineZhengzhou 450046China Department of Pharmacy Chongqing University Cancer HospitalChongqing 400030China School of Pharmaceutical Sciences Henan UniversityZhengzhou 450046China

Postoperative tumor recurrence remains a predominant cause of treatment failure. In this study, we developed an in situ injectable hydrogel, termed MPB-NO@DOX + ATRA gel, which was locally formed within the tumor resection cavity. The MPB-NO@DOX + ATRA gel was fabricated by mixing a thrombin solution, a fibrinogen solution containing all-trans retinoic acid (ATRA), and a Mn/NO-based immune nano-activator termed MPB-NO@DOX. ATRA promoted the differentiation of cancer stem cells, inhibited cancer cell migration, and affected the polarization of tumor-associated macrophages. The outer MnO2 shell disintegrated due to its reaction with glutathione and hydrogen peroxide in the cytoplasm to release Mn2+ and produce O2, resulting in the release of doxorubicin (DOX). The released DOX entered the nucleus and destroyed DNA, and the fragmented DNA cooperated with Mn2+ to activate the cGAS-STING pathway and stimulate an anti-tumor immune response. In addition, when MPB-NO@DOX was exposed to 808 nm laser irradiation, the Fe-NO bond was broken to release NO, which downregulated the expression of PD-L1 on the surface of tumor cells and reversed the immunosuppressive tumor microenvironment. In conclusion, the MPB-NO@DOX + ATRA gel exhibited excellent anti-tumor efficacy. The results of this study demonstrated the great potential of in situ injectable hydrogels in preventing postoperative tumor recurrence.

关键词： Post-sur gical tumor recurrence In situl hydrogel Immunotherapy Tumor micr oenvir onment Manganese(Ⅱ) Nitic oxide

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

论哈贝马斯人与自然关系理论的现象学视域

自然辩证法通讯 2022年第8期44卷 35-42页

作者：雷德鹏李霞丽广西大学马克思主义学院广西南宁530004

哈贝马斯把现象学的“设计”和“兴趣”两个概念勾连起来而形成一种独特的哲学视域,以之审视人与自然的关系,既严厉地批判了近代以来在技术认识兴趣导向下把自然视为“开采对象”的狭隘的科学主义意识,又创造性地提出了应当以解放认识... 详细信息

哈贝马斯把现象学的“设计”和“兴趣”两个概念勾连起来而形成一种独特的哲学视域,以之审视人与自然的关系,既严厉地批判了近代以来在技术认识兴趣导向下把自然视为“开采对象”的狭隘的科学主义意识,又创造性地提出了应当以解放认识兴趣为导向把自然设计为人类的“生存伙伴”的新见解,从而构建起一种具有浓郁现象学气息的人与自然关系理论。

关键词：哈贝马斯认识兴趣人与自然关系现象学

A Multi-modal Searching Algorithm in Computer Go Based on Test

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

A Multi-modal Searching Algorithm in Computer Go Based on Te...

2015年中国智能自动化学术会议

作者： xiali li licheng Wu School of Information Engineering Minzu University of China

Pattern matching algorithms based on domain knowledge have a weak global sense. Monte Carlo algorithms have a weak tactical ability. This article proposed a multi-modal algorithm on the basis of tests. GNU Go software was used to make experiments. Through the analysis of the experiment data, we set the move thresholds of start, middle, and the end of a Go play game. In the start stage, pattern matching algorithm was used. In the middle stage, Monte Carlo tree search algorithm with pruning was used to search the optimal moves. In the end stage, pattern matching and crazy model were used. The multi-modal algorithm was applied to develop software. Experiments demonstrate that multi-modal algorithm can improve the performance of Go in 13 × 13 board and 19 × 19 board.

关键词： Go Pattern matching Monte Carlo Pruning Domain knowledge Multi-modal

P2P Network Trust Strategy Based on New Evaluation Criterion

在线全文

cnki会议

学校读者我要写书评

暂无评论

P2P Network Trust Strategy Based on New Evaluation Criterion

2015年中国智能自动化学术会议

作者： xiali li licheng Wu School of Information Engineering Minzu University of China

In this paper we proposed a new comprehensive trust strategy which uses Malicious Percentage(MP), Feedback Consistency Percentage(FCP) and Initial Trust Vector(IPV) as the new trust evaluation criterion to help filter malicious peers in selection of peers. Simulations demonstrate that the new trust strategy can speed the execution time of the algorithm and enhance the successful transaction rate when resisting cooperative-cheating peers.

关键词： P2P network Trust strategy Malicious percentage(MP) Feedback consistency percentage(FCP) Initial trust vector(IPV)

UCT Algorithm in Amazons Human-computer Games

在线全文

cnki会议

学校读者我要写书评

暂无评论

UCT Algorithm in Amazons Human-computer Games

第26届中国控制与决策会议

作者： xiali li liang Hou licheng Wu School of Information Engineering MINZU University of China

The main difficulty of Amazons games is the huge branch factor. Traditional NegaMax Search and Alpha-Beta Search can only search a few layers and can’t improve the games effectively. Monte-Carlo Algorithm can improve the games but it needs a huge amount of computing. This paper studies applying UCT algorithm to the Amazons games to overcome this problem in order to balance the search efficiency and computing load. In UCT mini-max tree search, the algorithm select tree node according to the node’s UCB(Upper Confidence Bound) *** evaluate the selected node and return the optimal moves. Using C++ programming language, we implemented Amazons human-computer games software. The experiments show that UCT algorithm can implement the search work in Amazons human-computer games and get satisfactory search efficiency.

关键词： Artificial Intelligence Computer Games Amazons UCT Algorithm Monte-Carlo

Review and Research Issues on Underactuated Finger Mechanism

在线全文

cnki会议

学校读者我要写书评

暂无评论

Review and Research Issues on Underactuated Finger Mechanism

2015年中国智能自动化学术会议

作者： licheng Wu Yanxuan Kong xiali li Minzu University of China

The current typical prototypes, design methods, and analysis theory on underactuated finger mechanism are reviewed. Primary existing problems are also pointed out according to some analyses. Finally, several important... 详细信息

关键词： Robotic hand Robotic finger Underactuated finger mechanism(UFM) The state of the art