Digital twin driven and intelligence enabled content delivery in end-edge-cloud collaborative 5G networks

The rapid development of 5G/6G and AI enables an environment of Internet of Everything(IoE)which can support millions of connected mobile devices and applications to operate smoothly at high speed and low delay.However,these massive devices will lead to explosive traffic growth,which in turn cause great burden for the data transmission and content delivery.This challenge can be eased by sinking some critical content from cloud to edge.In this case,how to determine the critical content,where to sink and how to access the content correctly and efficiently become new challenges.This work focuses on establishing a highly efficient content delivery framework in the IoE environment.In particular,the IoE environment is re-constructed as an end-edge-cloud collaborative system,in which the concept of digital twin is applied to promote the collaboration.Based on the digital asset obtained by digital twin from end users,a content popularity prediction scheme is firstly proposed to decide the critical content by using the Temporal Pattern Attention(TPA)enabled Long Short-Term Memory(LSTM)model.Then,the prediction results are input for the proposed caching scheme to decide where to sink the critical content by using the Reinforce Learning(RL)technology.Finally,a collaborative routing scheme is proposed to determine the way to access the content with the objective of minimizing overhead.The experimental results indicate that the proposed schemes outperform the state-of-the-art benchmarks in terms of the caching hit rate,the average throughput,the successful content delivery rate and the average routing overhead.

一种基于BiLSTM 的混合层次化图分类模型

图分类在化学和生物信息学等诸多领域中是一个非常重要且极具挑战的问题,GNN模型是图分类问题的主流方法。现有的GNN模型采用卷积操作来实现邻域节点信息聚集,再通过池化操作生成粗化图。然而,仅通过池化方法不能捕获到每次卷积后读出图的双向依赖关系。为了提取到更充分的特征信息,提出一种混合层次化模型,首先分别提取节点特征信息和结构特征信息,再将特征信息融合,然后采用BiLSTM捕获不同层次读出图之间的双向依赖关系,从而提取到更丰富的特征信息。实验结果表明,与对比模型相比,上述模型的准确度有着明显的提升。

为上下文显式独立建模的中文实体识别方法

现有中文命名实体识别(NER)模型在公开数据集上的表现相对成熟,但有研究指出,模型过度依赖实体文本的字面特征,而上下文对实体识别的影响却未得到重视。现有的模型在简单的泛化测试中表现较差,因此本文提出显式地为上下文独立建模,令模型对上下文和实体的字面信息进行区分。为此,也提出了相应的数据增强方法用于训练模型中的上下文模块、实体字面模块和综合模块。实验结果表明,本文提出的方法在不损失测试集识别效果的情况下,明显改善了模型在不变性测试中的表现,较基准模型其失败率降低了2.3%。

Ada-FFL:Adaptive computing fairness federated learning

As the scale of federated learning expands,solving the Non-IID data problem of federated learning has become a key challenge of interest.Most existing solutions generally aim to solve the overall performance improvement of all clients;however,the overall performance improvement often sacrifices the performance of certain clients,such as clients with less data.Ignoring fairness may greatly reduce the willingness of some clients to participate in federated learning.In order to solve the above problem,the authors propose Ada-FFL,an adaptive fairness federated aggregation learning algorithm,which can dynamically adjust the fairness coefficient according to the update of the local models,ensuring the convergence performance of the global model and the fairness between federated learning clients.By integrating coarse-grained and fine-grained equity solutions,the authors evaluate the deviation of local models by considering both global equity and individual equity,then the weight ratio will be dynamically allocated for each client based on the evaluated deviation value,which can ensure that the update differences of local models are fully considered in each round of training.Finally,by combining a regularisation term to limit the local model update to be closer to the global model,the sensitivity of the model to input perturbations can be reduced,and the generalisation ability of the global model can be improved.Through numerous experiments on several federal data sets,the authors show that our method has more advantages in convergence effect and fairness than the existing baselines.

Caching resource management of mobile edge network based on Stackelberg game

Mobile edge caching technology is gaining more and more attention because it can effectively improve the Quality of Experience (QoE) of users and reduce backhaul burden. This paper aims to improve the utility of mobile edge caching technology from the perspectie of caching resource management by examining a network composed of one operator, multiple users and Content Providers (CPs). The caching resource management model is constructed on the premise of fully considering the QoE of users and the servicing capability of the Base Station (BS). In order to create the best caching resource allocation scheme, the original problem is transformed into a multi-leader multi-follower Stackelberg game model through the analysis of the system model. The strategy combinations and the utility functions of players are analyzed. The existence and uniqueness of the Nash Equilibrium (NE) solution are also analyzed and proved. The optimal strategy combinations and the best responses are deduced in detail. Simulation results and analysis show that the proposed model and algorithm can achieve the optimal allocation of caching resource and improve the QoE of users.

Recent advances in deep learning based sentiment analysis

Sentiment analysis is one of the most popular research areas in natural language processing.It is extremely useful in many applications,such as social media monitoring and e-commerce.Recent application of deep learning based methods has dramatically changed the research strategies and improved the performance of many traditional sentiment analysis tasks,such as sentiment classification and aspect based sentiment analysis.Moreover,it also pushed the boundary of various sentiment analysis task,including sentiment classification of different text granularities and in different application scenarios,implicit sentiment analysis,multimodal sentiment analysis and generation of sentiment-bearing text.In this paper,we give a brief introduction to the recent advance of the deep learning-based methods in these sentiment analysis tasks,including summarizing the approaches and analyzing the dataset.This survey can be well suited for the researchers studying in this field as well as the researchers entering the field.

Multiple Impact Phenomenon in Impact Hammer Testing:Theoretical Analysis and Numerical Simulation

As conducting an impact hammer testing during experimental modal analysis,the multiple impact phenomenon must be avoided.It is generally recognized that the multiple impact phenomenon is induced by the tester’s improper operation and can be avoided through more careful operation.This study theoretically and numerically investigates the whole process of the dynamical interaction between the hammer tip and the impacted structure and discovers the intrinsically physical mechanism of the multiple impact phenomenon.The determination of the interacting process comes down to solve two sets of governing differential equations alternately,and the effectiveness of the theoretical analysis is validated by numerical simulations.Four dimensionless parameters governing the interacting process are recognized in the theoretical framework.The critical stiffness ratio for a given impacted location and the critical impacted location for a given stiffness ratio are analytically determined.These results can guide impact hammer testing to avoid the occurrence of multiple impact by suggesting the hammer tip and impacted locations.