Aim and Scope
 

Evolutionary computation(EC) is one of the most fruitful research areas in computation intelligence. Many evolutionary algorithms have been proposed and applied for continuous and combinatorial optimization problems in academic and industry. Especially in recent decades, adaptive evolutionary algorithms have been becoming which makes evolutionary algorithms more feasible to use and can easily handle diversity tasks. However, there are still many issues which are imperative to be addressed. First for an evolutionary algorithm, its control parameters are important and hard to configure as for different tasks different parameter should be set. Second, when optimizing a specific objective function, most of the adaptive evolutionary algorithms focus on how to use experience in current optimization trajectory (such as using current best solution to generate better solutions) but not considering using the experiences of optimizing some other objective functions. Last but not the least, most evolutionary algorithms are under several specific frameworks, no methods has ever been developed to adaptively adjust the frameworks.

Artificial Intelligence (AI) is one of the most attractive research areas. Machine Learning (ML) is a recognized, available way to realize artificial intelligence. Nowadays, machine learning based methods have been applied widely in various real scenarios. Recently, ML methods have been used for learning optimization algorithms. Learning to optimize aims to secure knowledge from the optimization experiences when optimizing some related objective functions and apply the knowledge for new but related functions. The `knowledge' should be about the components of an optimization algorithm. Up to now, `learning to optimize' has been used to learn classical optimization algorithms and have shown great potentials, but few on evolutionary algorithms. Research on `learning to optimize' for evolutionary algorithms can be a new research hotspots  in evolutionary computation.

 

Theme
 

The goal of this special session is to investigate how to use artificial intelligence method to improve the performance of evolutionary algorithms. This will deliver a snapshot of the latest advances in the contribution of artificial intelligence to the field of evolutionary computation. Topics of interest include, but are not limited to (the topic contains both single objective optimization problems and multi-objective optimization problems) the following:

•    AI-assisted evolutionary computation framework
•    AI-assisted adaptive evolutionary algorithms
•    AI-assisted co-evolution algorithms
•    AI-assisted parallelized and distributed realizations of evolutionary computation
•    Evolutionary computation assisted on AI for real-world applications
•    Convergence analysis of AI-assisted assistedevolutionary algorithms
•    AI-assisted multi-objective evolutionary computation
•    AI-assisted evolutionary algorithms for constrained optimization
•    AI-assisted meta-heuristics for combinatorial optimization

Here the AI methods can be any learning algorithms, such as classification algorithms, clustering algorithms, regression algorithms, deep learning, reinforcement learning, meta-learning, and many others.

Improtant Dates

• Paper Submission Deadline: 31 January 2021
• Paper Acceptance Notification: 22 March 2021
• Final Paper Submission Deadline: 7 April 2021
• Early Registration Deadline: 7 April 2021
• Conference Dates: 28 June 2021 – 1 July 2021

Organizers
 

​​​​​Jianyong Sun: School of Mathematics and Statistics, National Engineering Laboratory for Big Data Analytics, Xi'an Jiaotong University, Xi'an, China.
Email: jy.sun@xjtu.edu.cn

Qingfu Zhang: Department of Computer Science, City University of Hong Kong, Hong Kong.
Email: ​​​​​qingfu.zhang@cityu.edu.hk

Submissions
 

Papers for IEEE CEC 2021 should be submitted electronically through the Congress website at http://cec2021.mini.pw.edu.pl, and will be refereed by experts in the fields. For Special Session papers, please select the respective Special Session title under the list of research topics in the submission system. 

.