Self-Organization and Developments of Super-Individuals in Large Scale Systems: HoneyBees, Tetrahymena, Web data, Boids

Using powerful algorithms, we propose a new theory of collective intelligence by studying the dynamics of living and non-living populations and their collective phenomena. In particular, I will discuss two perspectives. 

First, we analyze a large amount of individual tracking data obtained from actual biological experiments (such as honeybees, Tetrahymena,,) to develop a new theory of emergent phenomena in collective motion, especially the macroscopic emergence phenomenon called the super-individual. A super-individual is a collective state created from (replicating) individuals that behave as if they were individuals. 

The second is to build a theory of collective intelligence by analyzing the collective motion of populations at scales of tens of millions of individuals through large scale computer experiments (e.g. Boids model). Collective intelligence here is the “vital” intelligence as a  behavior of a group of individuals, and we explore the possibility  in a large swarming dynamics.

     Figure 2: Snapshots of simulated swarming behavior of Boids model. The total number of boids is 524288. [1]

References

[1] Ikegami,T., Mototake, Y., Kobori,S., Oka, M., and Hashimoto,Y.: Life as an emergent phenomenon: studies from a large-scale boid simulation and web data, Phil.Roy.Soc.,375, pp.1-15, 2017.

[2]  Doi, D., & Ikegami, T.  "Endogenous and Exogenous Bursts in a Honey Bee Hive." Artificial Life Conference Proceedings. MIT Press,pp. 493-499. 2018.& preprint(2020)

[3] Maruyama, N., Saito, D., & Ikegami, T. (2019, December). Emergence of Superorganisms in a Large Scale Boids Model. In 2019 IEEE Symposium Series on Computational Intelligence (SSCI), 299-305, IEEE.

[4] Ikegami, T., Hashimoto, Y., Oka, M. Open-Ended Evolution and a Mechanism of Novelties in Web Services. Artificial Life, pp.168-177, 2019.

 

Συνεδρία: 
Authors: 
Takashi Ikegami
Room: 
1
Date: 
Friday, December 11, 2020 - 12:40 to 13:20

Partners

Twitter

Facebook

Contact

For information please contact :
ccs2020conf@gmail.com