Parameter adaptation in multifactorial evolutionary algorithm for many-task optimization

Thang, Ta Bao; Dao, Tran Cong; Long, Nguyễn Hoàng; Bình, Huỳnh Thị Thanh

doi:10.1007/s12293-021-00347-4

Cited by 11 publications

(4 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To summarize, a good design of an EMTO algorithm usually considers the issues of "what and how to transfer" and "when to transfer" simultaneously. In addition, there are also other studies that focus on different types of problems by EMTO algorithms, such as many-task optimization problems [34]- [37], combinatorial problems [38]- [40], multiobjective problems [41]- [43], computationally expensive problems [44]- [46], theoretical analysis of the effectiveness of the EMTO algorithm [47]- [49], and applications of EMTO algorithms to real-world problems [50]- [53].…”

Section: B Related Workmentioning

confidence: 99%

Orthogonal Transfer for Multitask Optimization

Zhan

Tan

et al. 2023

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

Knowledge transfer (KT) plays a key role in multitask optimization. However, most of the existing KT methods still face two challenges. First, the tasks may commonly have different dimensionalities, making the KT between heterogeneous search spaces very difficult. Second, the tasks may have different degrees of similarity in different dimensions, making that treating all dimensions with equal importance may be harmful to the KT process. To address these two challenges, this paper proposes a novel orthogonal transfer (OT) method that is enabled by a cross-task mapping (CTM) strategy, which can achieve high-quality KT among heterogeneous tasks. For the first challenge, the CTM strategy maps the global best individual of one task from its original search space to the search space of the target task via an optimization process, which can handle the difference in task dimensionality. For the second challenge, the OT method is performed on the CTM-obtained individual and a random individual of the target task to find the best combination of different dimensions in these two individuals rather than treating all the dimensions equally, so as to achieve high-quality KT. To verify the effectiveness of the proposed OT method and the resulted OT-based multitask optimization (OTMTO) algorithm, this paper not only uses the existing multitask optimization benchmark but also proposes a new benchmark test suite named multitask optimization problems with different dimensionalities. Comprehensive experimental results on the existing and the proposed benchmarks show that the proposed OT method and the OTMTO algorithm are very advantageous in providing high-quality KT and in handling the heterogeneity of search space in multitask optimization problems compared to the existing competitive evolutionary multitask optimization algorithms.

show abstract

Section: B Related Workmentioning

confidence: 99%

Orthogonal Transfer for Multitask Optimization

Zhan

Tan

et al. 2023

IEEE Trans. Evol. Computat.

View full text Add to dashboard Cite

show abstract

“…Evolutionary multi-task optimization (EMTO) [45][46][47][48][49] is an emerging paradigm in the field of evolutionary computation. By sharing searched knowledge in similar tasks, EMTO can improve the convergence characteristics and searching efficiency for each task [50].…”

Section: Evolutionary Multi-task Optimizationmentioning

confidence: 99%

Network Collaborative Pruning Method for Hyperspectral Image Classification Based on Evolutionary Multi-Task Optimization

et al. 2023

View full text Add to dashboard Cite

Neural network models for hyperspectral images classification are complex and therefore difficult to deploy directly onto mobile platforms. Neural network model compression methods can effectively optimize the storage space and inference time of the model while maintaining the accuracy. Although automated pruning methods can avoid designing pruning rules, they face the problem of search efficiency when optimizing complex networks. In this paper, a network collaborative pruning method is proposed for hyperspectral image classification based on evolutionary multi-task optimization. The proposed method allows classification networks to perform the model pruning task on multiple hyperspectral images simultaneously. Knowledge (the important local sparse structure of the network) is automatically searched and updated by using knowledge transfer between different tasks. The self-adaptive knowledge transfer strategy based on historical information and dormancy mechanism is designed to avoid possible negative transfer and unnecessary consumption of computing resources. The pruned networks can achieve high classification accuracy on hyperspectral data with limited labeled samples. Experiments on multiple hyperspectral images show that the proposed method can effectively realize the compression of the network model and the classification of hyperspectral images.

show abstract

“…RQ2: Which factor is essential to the effectiveness of similarity metrics in S-ESTO? MTOP [13][14][15] Adaptation-based STOP [6,7,9,16,17] RQ3: How do existing adaptation techniques perform in solution adaptation for S-ESTO? RQ4: What contributes to the effectiveness of solution adaptation models in S-ESTO?…”

Section: Introductionmentioning

confidence: 99%

“…b In [15,29], inter-task similarity is measured by the feedback-relevant information instead of a feedforward metric.…”

Section: Introductionmentioning

confidence: 99%

How to Exploit Optimization Experience? Revisiting Evolutionary Sequential Transfer Optimization: Part B - Algorithm Analysis

Xue¹,

Yang²,

Feng³

et al. 2022

Preprint

View full text Add to dashboard Cite

<p>This paper is the second part of a two-part series on evolutionary sequential transfer optimization (ESTO). The first part designs a problem generator to generate benchmark problems with diverse properties that can better resemble real-world problems. This part proposes five research questions (RQs) to empirically revisit a large body of knowledge transfer techniques in the context of solution-based ESTO (S-ESTO). Particularly, knowledge transfer techniques are organized from a perspective of transfer strategy, including solution selection, solution adaptation, and the integration of them. The results reveal that the performance of many algorithms is highly sensitive to similarity distribution of STOPs, which suggest the necessity of more flexible algorithm design in future studies. Specifically, when an STOP is full of source tasks that are highly similar to the target task in terms of optimum, one can randomly select one source task and reuse its elite solution(s) to accelerate the target search significantly. When an STOP contains source tasks with mixed similarities to the target task, an effective similarity metric that can represent the transferability is able to identify highly transferable solution(s). When an STOP is full of source tasks with low similarities to the target task, the transferability is expected to be improved by solution adaptation. In solution adaptation, a similarity metric in the form of a maximization (or minimization) objective function that is responsible for representing the transferability is optimized to obtain a mapping function for adapting source solution(s). The source code for reproducing our experiments is available at https://github.com/XmingHsueh/Revisiting-S-ESTOs-PartB.</p>

show abstract

Parameter adaptation in multifactorial evolutionary algorithm for many-task optimization

Cited by 11 publications

References 22 publications

Orthogonal Transfer for Multitask Optimization

Orthogonal Transfer for Multitask Optimization

Network Collaborative Pruning Method for Hyperspectral Image Classification Based on Evolutionary Multi-Task Optimization

How to Exploit Optimization Experience? Revisiting Evolutionary Sequential Transfer Optimization: Part B - Algorithm Analysis

Contact Info

Product

Resources

About