多智能体强化学习的机械臂运动控制决策研究

针对传统运动控算法存在环境适应性较差,效率低的问题。可以利用强化学习在环境中不断去探索试错,并通过奖励函数对神经网络参数进行调节的方法对机械臂的运动进行控制。但是在现实中无法提供机械臂试错的环境,采用Unity引擎平台来构建机械臂的数字孪生仿真环境,设置观察状态变量和设置奖励函数机制,并提出在该模型环境中对PPO(proximal policy optimization)与多智能体（agents）结合的M-PPO算法来加快训练速度,实现通过强化学习算法对机械臂进行智能运动控制,完成机械臂执行末端有效避障快速到达目标物体位置,并通过该算法与M-SAC（多智能体与Soft Actor-Critic结合）和PPO算法的实验结果进行分析,验证M-PPO算法在不同环境下机械臂运动控制决策调试上的有效性与先进性。实现孪生体自主规划决策,反向控制物理体同步运动的目的。     

融合CNN-BiLSTM-Attention的集成学习价格预测

价格预测对于大宗农产品市场的稳定具有重要意义,但是大宗农产品价格与多种因素有着复杂的相关关系.针对当前价格预测中对数据完整性依赖性强与单一模型难以全面利用多种数据特征等问题,提出了一种将基于注意力机制的卷积双向长短期记忆神经网络(CNN-BiLSTM-Attention)、支持向量机回归(SVR)与LightGBM组合的增强式集成学习方法,并分别在包含历史交易、天气、汇率、油价等多种特征数据的数据集上进行了实验.实验以小麦和棉花价格预测为目标任务,使用互信息法进行特征选择,选择误差较低的CNN-BiLSTM-Attention模型作为基模型,与机器学习模型通过线性回归进行增强式集成学习.实验结果表明该集成学习方法在小麦及棉花数据集上预测结果的均方根误差(RMSE)值分别为12.812, 74.365,较之3个基模型分别降低11.00%, 0.94%、4.44%,1.99%与13.03%, 4.39%,能够有效降低价格预测的误差.     

区块链电子存证系统的设计与实现

文中介绍了区块链技术在电子证据存证方面的应用。通过将需要的电子证据以交易的形式记录下来,并明确记录时间戳,然后将记录存储在节点账本中。利用分布式数据存储技术完成电子存证过程,区块链技术可以实现不可篡改的信息存取和交换,有效保证电子数据的安全性和可信性。文中还探讨了区块链技术在降低处理电子证据时的技术风险和存储成本以及其对司法机构的协助作用。     

针对运动目标的多无人机协同鸽群优化搜索方法

针对多无人机协同运动目标搜索问题,本文设计了改进鸽群优化算法的协同搜索决策.首先,基于运动目标的独立性,建立了服从正态分布的目标概率信息图模型;为了提高环境中目标存在的确定度,建立了搜索环境的确定度信息图.其次,通过建立的吸引和排斥数字信息素图,引导无人机向未搜索区域飞行,减少重复搜索概率,提高协同目标搜索效率,并基于传统的鸽群算法,通过加入速度更新修正机制和精英代机制对其进行改进.然后,结合环境中目标的存在概率信息以及无人机搜索目标的探测信息,使用改进鸽群优化算法,规划无人机的最优搜索飞行路径.并设计避碰机制,以有效防止无人机搜索过程中的碰撞.最后,通过比较仿真实验验证了改进鸽群优化算法对运动目标协同搜索的有效性.     

Weakly supervised temporal action localization with proxy metric modeling

Temporal localization is crucial for action video recognition. Since the manual annotations are expensive and time-consuming in videos, temporal localization with weak video-level labels is challenging but indispensable. In this paper, we propose a weakly-supervised temporal action localization approach in untrimmed videos. To settle this issue, we train the model based on the proxies of each action class. The proxies are used to measure the distances between action segments and different original action features. We use a proxy-based metric to cluster the same actions together and separate actions from backgrounds. Compared with state-of-the-art methods, our method achieved competitive results on the THUMOS14 and ActivityNet1.2 datasets.     

Jointly beam stealing attackers detection and localization without training: an image processing viewpoint

Recently revealed beam stealing attacks could greatly threaten the security and privacy of IEEE 802.11ad communications. The premise to restore normal network service is detecting and locating beam stealing attackers without their cooperation. Current consistency-based methods are only valid for one single attacker and are parameter-sensitive. From the viewpoint of image processing, this paper proposes an algorithm to jointly detect and locate multiple beam stealing attackers based on RSSI (Received Signal Strength Indicator) map without the training process involved in deep learning-based solutions. Firstly, an RSSI map is constructed based on interpolating the raw RSSI data for enabling high-resolution localization while reducing monitoring cost. Secondly, three image processing steps, including edge detection and segmentation, are conducted on the constructed RSSI map to detect and locate multiple attackers without any prior knowledge about the attackers. To evaluate our proposal’s performance, a series of experiments are conducted based on the collected data. Experimental results have shown that in typical parameter settings, our algorithm’s positioning error does not exceed 0.41 m with a detection rate no less than 91%.     

Few-shot node classification via local adaptive discriminant structure learning

Node classification has a wide range of application scenarios such as citation analysis and social network analysis. In many real-world attributed networks, a large portion of classes only contain limited labeled nodes. Most of the existing node classification methods cannot be used for few-shot node classification. To train the model effectively and improve the robustness and reliability of the model with scarce labeled samples, in this paper, we propose a local adaptive discriminant structure learning (LADSL) method for few-shot node classification. LADSL aims to properly represent the nodes in the attributed graphs and learn a metric space with a strong discriminating power by reducing the intra-class variations and enlarging inter-class differences. Extensive experiments conducted on various attributed networks datasets demonstrate that LADSL is superior to the other methods on few-shot node classification task.     

“互联网＋”政务大数据智能服务平台设计与应用

“互联网＋”环境下,政务大数据关联公共服务和社会传感数据,使城市管理模式从单一走向立体、城市服务系统从孤立走向共享、城市决策模式从机械走向智能,已成为当前政务服务发展趋势。为解决政务数据、公共服务数据和社会传感数据融合不足、共享不充分、利用效能不高及技术平台缺乏等问题,该文提出了“互联网＋”政务大数据智能服务平台总体框架,首先对平台建设理论进行阐述,其次对系统架构、技术架构、关键技术和服务应用等进行设计,最后实现“互联网＋”政务大数据智能服务平台透明访问、城市信息单元建模、透明智能体协同、透明管理和微服务等功能。同时,该平台关键技术在数字广东和辽宁丹东社会保障等平台进行探索性应用,结果显示该平台能有效支持政务数据、公共服务数据和社会传感数据高效共享、管理和利用,提升部门服务效能。该文提出的平台总体设计方案可为省市级政府进行“互联网＋”政务大数据智能服务建设提供有价值的参考。     

VIS+AI: integrating visualization with artificial intelligence for efficient data analysis

Visualization and artificial intelligence (AI) are well-applied approaches to data analysis. On one hand, visualization can facilitate humans in data understanding through intuitive visual representation and interactive exploration. On the other hand, AI is able to learn from data and implement bulky tasks for humans. In complex data analysis scenarios, like epidemic traceability and city planning, humans need to understand large-scale data and make decisions, which requires complementing the strengths of both visualization and AI. Existing studies have introduced AI-assisted visualization as AI4VIS and visualization-assisted AI as VIS4AI. However, how can AI and visualization complement each other and be integrated into data analysis processes are still missing. In this paper, we define three integration levels of visualization and AI. The highest integration level is described as the framework of VIS+AI, which allows AI to learn human intelligence from interactions and communicate with humans through visual interfaces. We also summarize future directions of VIS+AI to inspire related studies.