posted by organizer: atzanetos || 835 views || tracked by 1 users: [display]

NIA2NIA CEC 2021 : Special Session on CEC 2021: Nature-Inspired Intelligence at the service of Nature-Inspired Intelligence


When N/A
Where Kraków, Poland
Submission Deadline Jan 31, 2021
Notification Due Mar 22, 2021
Final Version Due Apr 7, 2021
Categories    nature-inspired algorithms   real-world problems   premature convergence   hybridization and adaptation

Call For Papers

The last decade, the number of new stochastic nature-inspired algorithms has grown exponentially. The majority of these algorithms does not consist of a new optimization approach, but seems a lot similar to established ones, like Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO) or Differential Evolution (DE). In contrast, a few studies noted this fact and suggested to hybridize and adapt the existing techniques instead of looking for a new inspiration. On that basis, a good practice would be to use the mechanisms of an algorithm to enhance the performance or to overcome the drawbacks of another one. For example, the scatter mechanism of Mine Blast Algorithm could be used to enhance diversification of solutions of PSO, in order to overcome trapping in local optima.

This Special Session focuses on the investigation of mechanisms included in stochastic nature-inspired (or meta-heuristic) algorithms that could be used to modify the algorithmic process of another nature-inspired algorithm. The aim is to enhance the performance of existing nature-inspired algorithms and also, overcome drawbacks of the Vanilla algorithm, such as premature convergence and local optima trapping.

Furthermore, the proposal of such a modification to produce an even better method to solve a real-world application is encouraged. However, testing on CEC benchmark problems and functions is legitimate. In any case, non-parametric tests and appropriate tests to support the initial research hypothesis are needed.

This Special Session considers Nature-Inspired Algorithms based on the definition given by (K. C. B. Steer, A. Wirth, and S. K. Halgamuge, 2009): “The term nature refers to any part of the physical universe which is not a product of intentional human design”. Proposals of new methods are not encouraged and also, the study of algorithms that do not fall within the above definition (e.g. social-inspired approaches) should be avoided.

Topics of interest include, but are not limited to:
- Implementation of mechanisms that are part of nature-inspired algorithms
- Hybridization and adaptation of algorithms
- Premature convergence
- Exploration/exploitation balance
- Applications of modified nature-inspired on real-world problems
- Theoretical analysis of new nature-inspired algorithms

Related Resources

CEC 2021   IEEE CEC Special Session on Large Scale Global Optimization
CEC 2021   IEEE Congress on Evolutionary Computation
CEC-RIAEC 2021   IEEE CEC-SS Real-World and Industry Applications of Evolutionary Computation
EAI ICTCC 2021   7th EAI International Conference on Nature of Computation and Communication (Springer, Scopus, ISI, Ei, more)
ICANN 2021   30th International Conference on Artificial Neural Networks
Bio-inspired Deep Learning 2021   CFP: Bio-inspired Deep Learning Image and Signal Processing Pipelines in Medical Oncology - PeerJ
WCIE 2021   4th World Congress on Inspired Education
KES 2021   KES2021: Decision support systems, forecasting, classification - invited session IS14
EAI BICT 2021   13th EAI International Conference on Bio-inspired Information and Communications Technologies (Springer, Scopus, ISI, Ei, more)
Scopus-ASB 2021   2021 Asian Symposium on Biorobotics (ASB 2021)