Some Internet-of-Things (IoT) applications have a strict requirement on the end-to-end delay where edge computing can be used to provide a short delay for end-users by conducing efficient caching and computing at the edge nodes. However, a fast and efficient communication route creation in multi-access vehicular environment is an underexplored research problem. In this paper, we propose a collaborative learning-based routing scheme for multi-access vehicular edge computing environment. The proposed scheme employs a reinforcement learning algorithm based on end-edge-cloud collaboration to find routes in a proactive manner with a low communication overhead. The routes are also preemptively changed based on the learned information. By integrating the “proactive” and “preemptive” approach, the proposed scheme can achieve a better forwarding of packets as compared with existing alternatives. We conduct extensive and realistic computer simulations to show the performance advantage of the proposed scheme over existing baselines.

/pdf/collaborative-learning-of-communication-routes-in-edge-2s5fo73kt3.pdf

Collaborative Learning of Communication Routes in Edge-Enabled Multi-Access Vehicular Environment

Computation offloading from a mobile device to the edge server is an emerging paradigm to reduce completion latency of intensive computations in mobile-edge computing (MEC). In order to satisfy the delay-sensitive computing tasks, offloading time, including task uploading time, task execution time, and results downloading time is adopted as the computational performance metrics for offloading nodes that perform offloaded computing tasks for mobile devices. Therefore, how to minimize the offloading time by selecting an optimal offloading node in MEC is of research importance. This work first investigates a MEC system consisting of mobile devices and heterogeneous edge severs that support various radio access technologies. Then, based on the available bandwidth of heterogeneous edge severs and the location of mobile devices, an optimal offloading node selection strategy is formulated as a Markov decision process (MDP), and solved by employing the value iteration algorithm (VIA). Finally, extensive numerical results demonstrate the effectiveness of the proposed strategy over classic strategies in terms of offloading time.

Offloading Time Optimization via Markov Decision Process in Mobile-Edge Computing

With significant advances in communication and computing, modern day vehicles are becoming increasingly intelligent. This gives them the ability to contribute to safer roads and passenger comfort through network devices, cameras, sensors, and computational storage and processing capabilities. However, to run new and popular applications, and to enable vehicles operating autonomously requires massive computational resources. Computational resources available with the current day vehicles are not sufficient to process all these demands. In this situation, other vehicles, edge servers, and servers in remote data centers can help the vehicles by lending their computing resources. However, to take advantage of these computing resources, computation offloading techniques have to be leveraged to transfer tasks or entire applications to run on other devices. Such computation offloading can lead to improved performance and Quality of Service (QoS) for applications and for the network. However, computation offloading in a highly dynamic environment such as vehicular networks is a major challenge. Therefore, this survey aims to review and organize the computation offloading literature in vehicular environments. In addition, we demystify some concepts, propose a taxonomy with the most important aspects and classify most works in the area according to each category. We also present the main tools, scenarios, subjects, strategies, objectives, etc., used in the works. Finally, we present the main challenges and future directions to guide future research in this active research area.

/pdf/computation-offloading-for-vehicular-environments-a-survey-2w15qtgmxd.pdf

Computation Offloading for Vehicular Environments: A Survey

This paper investigates computing resource scheduling for real-time applications in autonomous driving, such as localization and obstacle avoidance. In our considered scenario, autonomous vehicles periodically sense the environment, offload sensor data to an edge server for processing, and receive computing results from the server. Due to mobility and computing latency, a vehicle travels some distance in the duration between the instant of offloading its sensor data and the instant of receiving the computing result. Our objective is finding a scheduling scheme for the edge sever to minimize the above traveled distance of vehicles. The approach is to determine the processing order according to individual vehicle mobility and computing capability of the edge server. We formulate a restless multi-arm bandit (RMAB) problem, design a Whittle index based stochastic scheduling scheme, and determine the index using a deep reinforcement learning (DRL) method. The proposed scheduling scheme avoids the time-consuming policy exploration common in DRL scheduling approaches and makes effectual decisions with low complexity. Extensive simulation results demonstrate that the proposed indexed-based scheme can deliver computing results to the vehicles promptly while adapting to time-variant vehicle mobility.

Adaptive Computing Scheduling for Edge-Assisted Autonomous Driving

Collaborative edge computing (CEC) is a recent popular paradigm where different edge devices collaborate by sharing data and computation resources. One of the fundamental issues in CEC is to make task offloading decision. However, it is a challenging problem to solve as tasks can be offloaded to a device at multi-hop distance leading to conflicting network flows due to limited bandwidth constraint. There are some works on multi-hop computation offloading problem in the literature. However, existing works have not jointly considered multi-hop partial computation offloading and network flow scheduling that can cause network congestion and inefficient performance in terms of completion time. This article formulates the joint multi-task partial computation offloading and network flow scheduling problem to minimize the average completion time of all tasks. The formulated problem optimizes several dependent decision variables including partial offloading ratio, remote offloading device, start time of tasks, routing path, and start time of network flows. The problem is formulated as an MINLP optimization problem and shown to be NP-hard. We propose a joint partial offloading and flow scheduling heuristic (JPOFH) that decides partial offloading ratio by considering both waiting times at the devices and start time of network flows. We also do the relaxation of formulated MINLP problem to an LP problem using McCormick envelope to give a lower bound solution. Performance comparison done using simulation shows that JPOFH leads to up to 32 percent improvement in average completion time compared to benchmark solutions which do not make a joint decision.

Multi-Hop Multi-Task Partial Computation Offloading in Collaborative Edge Computing

Vehicular edge computing (VEC) offers a new paradigm to improve vehicular services and augment the capabilities of vehicles. In this article, we study the problem of task scheduling in VEC, where multiple computation-intensive vehicular applications can be offloaded to roadside units (RSUs) and each application can be further divided into multiple tasks with task dependency. The tasks can be scheduled to different mobile-edge computing servers on RSUs for execution to minimize the average completion time of multiple applications. Considering the completion time constraint of each application and the processing dependency of multiple tasks belonging to the same application, we formulate the multiple tasks scheduling problem as an optimization problem that is NP-hard. To solve the optimization problem, we develop an efficient task scheduling algorithm. The basic idea is to prioritize multiple applications and prioritize multiple tasks so as to guarantee the completion time constraints of applications and the processing dependency requirements of tasks. The numerical results demonstrate that our proposed algorithm can significantly reduce the average completion time of multiple applications compared with benchmark algorithms.

Dependency-Aware Task Scheduling in Vehicular Edge Computing

The Chimp Optimization Algorithm (ChOA) is a heuristic algorithm proposed in recent years. It models the cooperative hunting behaviour of chimpanzee populations in nature and can be used to solve numerical as well as practical engineering optimization problems. ChOA has the problems of slow convergence speed and easily falling into local optimum. In order to solve these problems, this paper proposes a novel chimp optimization algorithm with refraction learning (RL-ChOA). In RL-ChOA, the Tent chaotic map is used to initialize the population, which improves the population’s diversity and accelerates the algorithm’s convergence speed. Further, a refraction learning strategy based on the physical principle of light refraction is introduced in ChOA, which is essentially an Opposition-Based Learning, helping the population to jump out of the local optimum. Using 23 widely used benchmark test functions and two engineering design optimization problems proved that RL-ChOA has good optimization performance, fast convergence speed, and satisfactory engineering application optimization performance.

/pdf/a-novel-chimp-optimization-algorithm-with-refraction-2fpt1yos.pdf

A Novel Chimp Optimization Algorithm with Refraction Learning and Its Engineering Applications

Entity linking and predicate matching are two core tasks in the Chinese Knowledge Base Question Answering (CKBQA). Compared with the English entity linking task, the Chinese entity linking is extremely complicated, making accurate Chinese entity linking difficult. Meanwhile, strengthening the correlation between entities and predicates is the key to the accuracy of the question answering system. Therefore, we put forward a Bidirectional Encoder Representation from Transformers and transfer learning Knowledge Base Question Answering (BAT-KBQA) framework, which is on the basis of feature-enhanced Bidirectional Encoder Representation from Transformers (BERT), and then perform a Named Entity Recognition (NER) task, which is appropriate for Chinese datasets using transfer learning and the Bidirectional Long Short-Term Memory-Conditional Random Field (BiLSTM-CRF) model. We utilize a BERT-CNN (Convolutional Neural Network) model for entity disambiguation of the problem and candidate entities; based on the set of entities and predicates, a BERT-Softmax model with answer entity predicate features is introduced for predicate matching. The answer ultimately chooses to integrate entities and predicates scores to determine the definitive answer. The experimental results indicate that the model, which is developed by us, considerably enhances the overall performance of the Knowledge Base Question Answering (KBQA) and it has the potential to be generalizable. The model also has better performance on the dataset supplied by the NLPCC-ICCPOL2016 KBQA task with a mean F1 score of 87.74% compared to BB-KBQA.

https://www.mdpi.com/2076-3417/12/22/11522/pdf?version=1669098586

Enhancement of Question Answering System Accuracy via Transfer Learning and BERT

The slime mold algorithm (SMA) is a swarm-based metaheuristic algorithm inspired by the natural oscillatory patterns of slime molds. Compared with other algorithms, the SMA is competitive but still suffers from unbalanced development and exploration and the tendency to fall into local optima. To overcome these drawbacks, an improved SMA with a dynamic quantum rotation gate and opposition-based learning (DQOBLSMA) is proposed in this paper. Specifically, for the first time, two mechanisms are used simultaneously to improve the robustness of the original SMA: the dynamic quantum rotation gate and opposition-based learning. The dynamic quantum rotation gate proposes an adaptive parameter control strategy based on the fitness to achieve a balance between exploitation and exploration compared to the original quantum rotation gate. The opposition-based learning strategy enhances population diversity and avoids falling into the local optima. Twenty-three benchmark test functions verify the superiority of the DQOBLSMA. Three typical engineering design problems demonstrate the ability of the DQOBLSMA to solve practical problems. Experimental results show that the proposed algorithm outperforms other comparative algorithms in convergence speed, convergence accuracy, and reliability.

/pdf/improved-slime-mold-algorithm-with-dynamic-quantum-rotation-2t1lxqcj.pdf

Improved Slime Mold Algorithm with Dynamic Quantum Rotation Gate and Opposition-Based Learning for Global Optimization and Engineering Design Problems

The traditional grey wolf algorithm is widely used for feature selection. However, within complex feature multi-dimensional problems, the grey wolf algorithm is prone to reach locally optimal solutions and premature convergence. In this paper, a threshold binary grey wolf optimizer based on multi-elite interaction for feature selection (MTBGWO) is proposed. Firstly, the multi-population topology is adopted to enhance the population’s diversity for improving search space utilization. Secondly, an information interaction learning strategy is adopted for the update of sub-population elite wolf position (optimal position) via learning better position from other elite wolves; in order to improve the local exploitation ability of the sub-population. At the same time, the command of $\beta $ and $\delta $ wolves (second and third best positions) for population position updates is removed. Finally, a threshold approach is employed to convert the continuous position of grey wolf individuals into binary one to apply in the feature selection problem. Further, The MTBGWO algorithm proposed in this paper is compared with the traditional binary grey wolf algorithm (BGWO), binary whale algorithm (BWOA), as well as some recently developed novel algorithms to exhibit its superiority and robustness. Totally 16 classification datasets, from the UCI Machine Learning Repository, are chosen for comparison. The Wilcoxon’s rank-sum non-parametric statistical test is carried out at 5% significance level to evaluate whether the results of the proposed algorithms significantly differs from those of the other algorithms. In the experimental results for all datasets, the overall average accuracy of the MTBGWO algorithm is 94.7%, while the highest of the other algorithms is 92.8% and the selected feature subset is 25% of the total dataset. The MTBGWO algorithm selects much smaller subset of features than other algorithms. In terms of computational efficiency, the overall processing time of MTBGWO is 24.2 seconds, whereas HSGW is 44.1 seconds. The results reveal that the MTBGWO has shown its superiority in solving the feature selection problem.

Shiyu Du

Papers

Dependency-Aware Task Scheduling in Vehicular Edge Computing

A Novel Chimp Optimization Algorithm with Refraction Learning and Its Engineering Applications

Enhancement of Question Answering System Accuracy via Transfer Learning and BERT

Improved Slime Mold Algorithm with Dynamic Quantum Rotation Gate and Opposition-Based Learning for Global Optimization and Engineering Design Problems

Threshold Binary Grey Wolf Optimizer Based on Multi-Elite Interaction for Feature Selection