ZCZhaoquan Cai

Zhaoquan Cai

*This is an algorithmically generated author record based on publicly available information.

Zhaoquan Cai has a total of 35 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Evolutionary Optimization of Energy Consumption and Makespan of Workflow Execution in Clouds

    by Lining Xing, Jun Li, Zhaoquan Cai, Feng Hou

    Making sound trade-offs between the energy consumption and the makespan of workflow execution in cloud platforms remains a significant but challenging issue. So far, some works balance workflows’ energy consumption and makespan by adopting multi-objective evolutionary algorithms, but they often regard this as a black-box problem, resulting in the low efficiency of the evolutionary search. To compensate for the shortcomings of existing works, this paper mathematically formulates the cloud workflow scheduling for an infrastructure-as-a-service (IaaS) platform as a multi-objective optimization problem. Then, this paper tailors a knowledge-driven energy- and makespan-aware workflow scheduling algorithm, namely EMWSA. Specifically, a critical task adjustment-based local search strategy is proposed to intelligently adjust some critical tasks to the same resource of their successor tasks, striving to simultaneously reduce workflows’ energy consumption and makespan. Further, an idle gap reuse strategy is proposed to search the optimal energy consumption of each non-critical task without affecting the operation of other tasks, so as to further reduce energy consumption. Finally, in the context of real-world workflows and cloud platforms, we carry out comparative experiments to verify the superiority of the proposed EMWSA by significantly outperforming 4 representative baselines on 19 out of 20 workflow instances.


  • Open Access

    Hybrid Learning Moth Search Algorithm for Solving Multidimensional Knapsack Problems

    by Yanhong Feng, Hongmei Wang, Zhaoquan Cai, Mingliang Li, Xi Li

    The moth search algorithm (MS) is a relatively new metaheuristic optimization algorithm which mimics the phototaxis and Lévy flights of moths. Being an NP-hard problem, the 0–1 multidimensional knapsack problem (MKP) is a classical multi-constraint complicated combinatorial optimization problem with numerous applications. In this paper, we present a hybrid learning MS (HLMS) by incorporating two learning mechanisms, global-best harmony search (GHS) learning and Baldwinian learning for solving MKP. (1) GHS learning guides moth individuals to search for more valuable space and the potential dimensional learning uses the difference between two random dimensions to generate a large jump. (2) Baldwinian learning guides moth individuals to change the search space by making full use of the beneficial information of other individuals. Hence, GHS learning mainly provides global exploration and Baldwinian learning works for local exploitation. We demonstrate the competitiveness and effectiveness of the proposed HLMS by conducting extensive experiments on 87 benchmark instances. The experimental results show that the proposed HLMS has better or at least competitive performance against the original MS and some other state-of-the-art metaheuristic algorithms. In addition, the parameter sensitivity of Baldwinian learning is analyzed and two important components of HLMS are investigated to understand their impacts on the performance of the proposed algorithm.


  • Open Access

    A Two-State Dynamic Decomposition-Based Evolutionary Algorithm for Handling Many-Objective Optimization Problems

    by Lining Xing, Jun Li, Zhaoquan Cai, Feng Hou

    Decomposition-based many-objective evolutionary algorithms (D-MaOEAs) are brilliant at keeping population diversity for predefined reference vectors or points. However, studies indicate that the performance of an D-MaOEA strongly depends on the similarity between the shape of the reference vectors (points) and that of the PF (a set of Pareto-optimal solutions symbolizing balance among objectives of many-objective optimization problems) of the many-objective problem (MaOP). Generally, MaOPs with expected PFs are not realistic. Consequently, the inevitable weak similarity results in many inactive subspaces, creating huge difficulties for maintaining diversity. To address these issues, we propose a two-state method to judge the decomposition status according to the number of inactive reference vectors. Then, two novel reference vector adjustment strategies, set as parts of the environmental selection approach, are tailored for the two states to delete inactive reference vectors and add new active reference vectors, respectively, in order to ensure that the reference vectors are as close as possible to the PF of the optimization problem. Based on the above strategies and an efficient convergence performance indicator, an active reference vector-based two-state dynamic decomposition-base MaOEA, referred to as ART-DMaOEA, is developed in this paper. Extensive experiments were conducted on ART-DMaOEA and five state-of-the-art MaOEAs on MaF1-MaF9 and WFG1-WFG9, and the comparative results show that ART-DMaOEA has the most competitive overall performance.


  • Retracted

    RETRACTED: Improved multi-objective neural network to the complex investment decision-making evaluation

    by Li Shen, Zhaoquan Cai

    As the investment directions and importance of high-tech enterprises increased largely, it is especially important to choose the most appropriate and effective investment for high-tech enterprises according to national conditions and economic conditions. In view of this, this paper proposes and constructs an investment decision-making scheme evaluation method of high-tech enterprises based on multi-objective neural network. The feasibility of the method was verified by investment decision-making scheme evaluation example of commercial bank. The results show that investment decision-making scheme B can best balance initial investment and capacity elasticity. The proposed method can be generalized to evaluation of other similar investment decision-making schemes.


  • A multi-objective multicast routing optimization based on differential evolution in MANET

    by Wenhong Wei, Yong Qin, Zhaoquan Cai

    Purpose: The purpose of this paper is to propose a multi-objective differential evolution algorithm named as MOMR-DE to resolve multicast routing problem. In mobile ad hoc network (MANET), multicast routing is a non-deterministic polynomial -complete problem that deals with the various objectives and constraints. Quality of service (QoS) in the multicast routing problem mainly depends on cost, delay, jitter and bandwidth. So the cost, delay, jitter and bandwidth are always considered as multi-objective for designing multicast routing protocols. However, mobile node battery energy is finite and the network lifetime depends on node battery energy. If the battery power consumption is high in any one of the nodes, the chances of network’s life reduction due to path breaks are also more. On the other hand, node’s battery energy had to be consumed to guarantee high-level QoS in multicast routing to transmit correct data anywhere and at any time. Hence, the network lifetime should be considered as one objective of the multi-objective in the multicast routing problem. Design/methodology/approach: Recently, many metaheuristic algorithms formulate the multicast routing problem as a single-objective problem, although it obviously is a multi-objective optimization problem. In the MOMR-DE, the network lifetime, cost, delay, jitter and bandwidth are considered as five objectives. Furthermore, three QoS constraints which are maximum allowed delay, maximum allowed jitter and minimum requested bandwidth are included. In addition, we modify the crossover and mutation operators to build the shortest-path multicast tree to maximize network lifetime and bandwidth, minimize cost, delay and jitter. Findings: Two sets of experiments are conducted and compared with other algorithms for these problems. The simulation results show that our proposed method is capable of achieving faster convergence and is more preferable for multicast routing in MANET. Originality/value: In MANET, most metaheuristic algorithms formulate the multicast routing problem as a single-objective problem. However, this paper proposes a multi-objective differential evolution algorithm to resolve multicast routing problem, and the proposed algorithm is capable of achieving faster convergence and more preferable for multicast routing.


  • Feature extraction with triplet convolutional neural network for content-based image retrieval

    by Zhiyin Cai, Wei Gao, Zhuliang Yu, Jinhong Huang, Zhaoquan Cai

    Feature representations and similarity metric are important for content-based image retrieval (CBIR) tasks. Inspired by recent successes of convolutional neural network (CNN) with hierarchical features, in this paper, we apply a triplet convolutional neural network (Triplet-CNN) to learn features with the criterion of similarity metric. Furthermore, we propose some strategies to improve the performance of Triplet-CNN for CBIR tasks. In training stage, in order to avoid the problem of lacking data in training Triplet-CNN, we utilize dataset in similar domains to retrain the network. In feature extraction stage, since the rectified linear unit (ReLU) activation function drops all negative neurons, we extract the non-activation based features to preserve the information. And then we combine features from different layers to retrieve images. Our experimental results demonstrate that our method can improve the retrieval performance of CBIR tasks.


  • Open Access

    Batch image alignment via subspace recovery based on alternative sparsity pursuit

    by Xianhui Lin, Zhu Liang Yu, Zhenghui Gu, Jun Zhang, Zhaoquan Cai

    The problem of robust alignment of batches of images can be formulated as a low-rank matrix optimization problem, relying on the similarity of well-aligned images. Going further, observing that the images to be aligned are sampled from a union of low-rank subspaces, we propose a new method based on subspace recovery techniques to provide more robust and accurate alignment. The proposed method seeks a set of domain transformations which are applied to the unaligned images so that the resulting images are made as similar as possible. The resulting optimization problem can be linearized as a series of convex optimization problems which can be solved by alternative sparsity pursuit techniques. Compared to existing methods like robust alignment by sparse and low-rank models, the proposed method can more effectively solve the batch image alignment problem, and extract more similar structures from the misaligned images.