Emergent Technologies Task Force on

Collaborative Learning and Optimization

Chairs
Maoguo Gong

Collaborative Innovation Center for Computational Intelligence (OMEGA), Xidian University, Xi'an, China

Vice Chairs

Kai Qin

School of Computer Science and Software Engineering, Swinburne University of Technology, Melbourne, Australia

Ke Tang

School of Computer Science and Technology, University of Science and Technology of China, China

Members

• El-ghazali Talbi, University of Lille, France

• Florence Forbes, INRIA, France

• Gabriela Ochoa, University of Stirling, UK

• Jose A. Lozano, University of the Basque Country, Spain

• Jing Liang, Zhengzhou University, China

• Marc Schoenauer, INRIA, France

• Swagatam Das, Indian Statistical Institute, India

• Qingfu Zhang, City University of Hong Kong, Hong Kong

• Shan He, University of Birmingham, UK

• Thomas Bartz-Beielstein, Cologne University of Applied Science, Germany

• Thomas Jansen, Aberystwyth University, UK

• Xiaodong Li, RMIT University, Australia

• Yaochu Jin, University of Surrey, UK

• Zexuan Zhu, Shenzhen University, China

• Yew-Soon Ong, Nanyang Technological University

• Liang Feng, Chongqing University

Motivation

Learning and optimization are two essential tasks that computational intelligence aims to address. Numerous techniques have been developed for these two purposes separately. In fact, learning and optimization are closely related. On the one hand, learning can be formulated as a model-centric or data-centric optimization problem, and accordingly solved by optimization techniques. On the other hand, optimization can be regarded as an adaptive learning process, and thus tackled by learning methodologies. Moreover, learning techniques play a crucial role in competence analysis of optimizers, and meanwhile optimization techniques provide an effective way to analyze learners’ domains of competence.

Recent years have seen remarkable attempts at collaborative learning and optimization. For instance, learning classifier systems, evolutionary neural networks, evolutionary ensemble learning, evolutionary kernel machines, evolutionary clustering,  evolutionary data generation and extraction for learning, supervised classifiers’ domain of competence analysis using evolutionary multi-objective optimization, theoretical analysis of optimization techniques using learning theory, optimization by building and using probabilistic models, self-adaptive and tuning-free optimization, ensemble optimization, statistical analysis of evolutionary computation, automatic heuristic design, etc. These research efforts have led to a great deal of cutting-edge techniques in the corresponding research fields.

Nowadays, the emergence of more and more complex problems in real-world applications insistently calls for in-depth investigations of synergy between learning and optimization. Moreover, feasibility of implementing collaborative learning and optimization techniques on massive parallel systems must be seriously taken into account to ensure that large-scale problems can be solved in a reasonable time.

Goals

The primary goal of this task force is to promote research on collaborative learning and optimization. Moreover, this task force shall provide a forum for academic and industrial researchers from both learning and optimization communities to collaboratively explore promising directions of synergy between techniques from these two areas.

Scope

The scope of this task force covers, but is not limited to:

• Learning based on data-centric optimization

• Learning based on model-centric optimization

• Competence analysis of learners using optimization techniques

• Automatic selection of learning techniques for solving real-world problems

• Theoretical relationship between learning and optimization

• Optimization based on state-of-the-art learning techniques

• Self-adaptive and tuning-free optimization based on novel learning models

• Competence analysis of optimizers using statistical & machine learning techniques

• Automatic selection of optimization techniques for solving real-world problems

• Collaborative learning and optimization on massively parallel systems