Knowledge Reasoning Method Based on Deep Transfer Reinforcement Learning:DTRLpath

In recent years,with the continuous development of deep learning and knowledge graph reasoning methods,more and more researchers have shown great interest in improving knowledge graph reasoning methods by inferring missing facts through reasoning.By searching paths on the knowledge graph and making fact and link predictions based on these paths,deep learning-based Reinforcement Learning(RL)agents can demonstrate good performance and interpretability.Therefore,deep reinforcement learning-based knowledge reasoning methods have rapidly emerged in recent years and have become a hot research topic.However,even in a small and fixed knowledge graph reasoning action space,there are still a large number of invalid actions.It often leads to the interruption of RL agents’wandering due to the selection of invalid actions,resulting in a significant decrease in the success rate of path mining.In order to improve the success rate of RL agents in the early stages of path search,this article proposes a knowledge reasoning method based on Deep Transfer Reinforcement Learning path(DTRLpath).Before supervised pre-training and retraining,a pre-task of searching for effective actions in a single step is added.The RL agent is first trained in the pre-task to improve its ability to search for effective actions.Then,the trained agent is transferred to the target reasoning task for path search training,which improves its success rate in searching for target task paths.Finally,based on the comparative experimental results on the FB15K-237 and NELL-995 datasets,it can be concluded that the proposed method significantly improves the success rate of path search and outperforms similar methods in most reasoning tasks.

Introduction to the Special Issue on ComputerModeling for Smart Cities Applications

A significant fraction of the world’s population is living in cities. With the rapid development ofinformation and computing technologies (ICT), cities may be made smarter by embedding ICT intotheir infrastructure. By smarter, we mean that the city operation will be more efficient, cost-effective,energy-saving, be more connected, more secure, and more environmentally friendly. As such, a smartcity is typically defined as a city that has a strong integration with ICT in all its components, includingits physical components, social components, and business components [1,2].

Facile synthesis of UCNPs/Zn_xCd_(1-x)S nanocomposites excited by near-infrared light for photochemical reduction and removal of Cr(Ⅵ)

Photocatalysis driven by near-infrared（NIR）light is of scientific and technological interest for ex-ploiting solar energy.In this study,we demonstrate a facile hydrothermal process to synthesize core-shell nanoparticles combining upconversion nanoparticles（UCNPs）and alloyed ZnxCwhich can be excited using NIR or visible light.Morphologies,phase,and chemical composition have been investigated using field-emission scanning electron microscopy,transmission electron mi-croscopy,X-ray diffraction analysis,and atomic absorption spectroscopy.Moreover,we found that amorphous TiO2 layers existing in the final samples play an important role in formation ofyolk-shell nanoparticles,which bind the as-prepared ZnxCnanoparticlescan be tuna-ble by adjusting the amount of the Cd and Zn source compounds.The photochemical reduction of Cr（Ⅵ）in water has been performed to study the photocatalytic performance under irradiation by NIR light or a simulated solar light,showing efficient photoreduction and Cr（Ⅵ）removal over the/TiO2 yolk-shell nanoparticles.The as-prepared UCNPs@ZnxC/TiO2 nanoparticles show excellent production of hydroxyl radicals,which are responsible for the photochemical reduction of Cr（Ⅵ）to Cr（Ⅲ）.This study will provide an alternative strategy for en-vironmental wastewater treatment,making full use of solar energy.

基于设备性能的Web3D动态实时光影云渲染系统

本文面向多种硬件平台提出了一套Web3D实时动态光影的协同式渲染系统,该系统把Web前端的硬件性能作为整个云渲染系统中光影渲染任务分配的关键因素.对于Web前端性能较强的硬件设备,系统分配复杂度较高的光影渲染任务给前端,相应的云后端的渲染负担则有所降低;反之,系统则分配复杂度较低的光影渲染任务给前端,相应的云后端承担大部分的渲染任务.在上述机制的引导下,该系统的前后端部署了4类关键的实时光影渲染算法,最终通过对算法运行帧率、算法所在设备的运行效率以及光影渲染结果等多种数据的分析,验证了部署的合理性.