Artificial Creativity of Evolutionary Swarm Systems

1 like•140 views

This document describes research on an artificial system that exhibits creative behavior through evolutionary swarm interactions. The system consists of particles that interact locally based on kinetic parameters and compete to transmit their "recipes" to other particles. Over many iterations, this leads to the spontaneous emergence of diverse stable object patterns in the environment. Subjecting the system to different levels of environmental perturbations influenced the size and color diversity of the evolved objects. The researcher argues this bottom-up approach offers a novel path to machine creativity without direct optimization of a global fitness function.

Science

Artificial Creativity of
Evolutionary Swarm Systems
Hiroki Sayama
Center for Collective Dynamics of Complex Systems
Binghamton University, State University of New York
sayama@binghamton.edu

https://www.oreilly.com/ideas/open-endedness-
the-last-grand-challenge-youve-never-heard-of
2

Swarm
Chemistry
http://bingweb.binghamton.edu/~sayama/SwarmChemistry/
3

i
• Particles in a continuous open 2-D space
• Kinetic interactions with local neighbors
• No capability to distinguish different types
Model Assumptions
4

Kinetic Parameters
(Assigned to each particle individually)
5

i
Cohesion Alignment Separation
97 * (226.76, 3.11, 9.61, 0.15, 0.88, 43.35, 0.44, 1.0 )
38 * ( 57.47, 9.99, 35.18, 0.15, 0.37, 30.96, 0.05, 0.31)
56 * ( 15.25, 13.58, 3.82, 0.3, 0.8, 39.51, 0.43, 0.65)
31 * (113.21, 18.25, 38.21, 0.62, 0.46, 15.78, 0.49, 0.61)
Recipe
6

Artificial Creativity of Evolutionary Swarm Systems

Recipe Competition Between Active Particles
Competition between active particles
97 * (226.76, 3.11, 9.61, 0.15, 0.88, 43.35, 0.44, 1.0)
38 * (57.47, 9.99, 35.18, 0.15, 0.37, 30.96, 0.05, 0.31)
56 * (15.25, 13.58, 3.82, 0.3, 0.8, 39.51, 0.43, 0.65)
31 * (113.21, 18.25, 38.21, 0.62, 0.46, 15.78, 0.49, 0.61)
67 * (216.35, 11.75, 7.7, 0.83, 0.97, 97.31, 0.02, 0.38)
29 * (254.64, 7.28, 7.0, 0.95, 0.11, 22.41, 0.43, 0.31)
13 * (105.4, 3.55, 5.24, 0.34, 0.18, 23.53, 0.39, 0.24)
competition
function
winner:
8

97 * (226.76, 3.11, 9.61, 0.15, 0.88, 43.35, 0.44, 1.0)
38 * (57.47, 9.99, 35.18, 0.15, 0.37, 30.96, 0.05, 0.31)
56 * (15.25, 13.58, 3.82, 0.3, 0.8, 39.51, 0.43, 0.65)
31 * (113.21, 18.25, 38.21, 0.62, 0.46, 15.78, 0.49, 0.61)
Mutations
Recipe transmission & mutation
75 * (216.35, 11.75, 7.7, 0.83, 0.97, 97.31, 0.02, 0.38)
29 * (254.64, 7.28, 7.0, 0.95, 0.11, 28.56, 0.43, 0.31)
13 * (105.4, 3.55, 5.24, 0.34, 0.18, 23.53, 0.39, 0.24)
9
67 * (216.35, 11.75, 7.7, 0.83, 0.97, 97.31, 0.02, 0.38)
29 * (254.64, 7.28, 7.0, 0.95, 0.11, 22.41, 0.43, 0.31)
13 * (105.4, 3.55, 5.24, 0.34, 0.18, 23.53, 0.39, 0.24)

Sayama, IEEE ALIFE (2011) Sayama & Wong, ECAL (2011)
Is This Really Creative?

Simulate It Much Longer
Detect Objects Automatically

Environmental Perturbations
• Competition function switched from
“majority-relative” to either “faster”
or “slower” for T steps in every 2000
steps in either half of the space
T = 50 (short, moderate)
T = 500 (long, severe)

Detailed Object Tracking (Object Size)
With moderate environmental perturbations
Objects tend
to get bigger
over time

Detailed Object Tracking (Object Size)
With moderate environmental perturbations With severe environmental perturbations
But it also depends on the
severity of perturbations

Detailed Object Tracking (Color Diversity)
With moderate environmental perturbations With severe environmental perturbations
Severe perturbations may increase
color entropy of objects

Results
https://youtu.be/BpN-DE3o6u8

Conclusions
Evolutionary swarm chemistry
can produce patterns with no
global fitness evaluation
The “bottom-up” approach
offers a unique path to
machine creativity
Automated object harvesting
has captured a large number
of patterns

Artificial Creativity of Evolutionary Swarm Systems

1. Artificial Creativity of Evolutionary Swarm Systems Hiroki Sayama Center for Collective Dynamics of Complex Systems Binghamton University, State University of New York [email protected]

2. https://www.oreilly.com/ideas/open-endedness- the-last-grand-challenge-youve-never-heard-of 2

3. Swarm Chemistry http://bingweb.binghamton.edu/~sayama/SwarmChemistry/ 3

4. i • Particles in a continuous open 2-D space • Kinetic interactions with local neighbors • No capability to distinguish different types Model Assumptions 4

5. Kinetic Parameters (Assigned to each particle individually) 5

6. i Cohesion Alignment Separation 97 * (226.76, 3.11, 9.61, 0.15, 0.88, 43.35, 0.44, 1.0 ) 38 * ( 57.47, 9.99, 35.18, 0.15, 0.37, 30.96, 0.05, 0.31) 56 * ( 15.25, 13.58, 3.82, 0.3, 0.8, 39.51, 0.43, 0.65) 31 * (113.21, 18.25, 38.21, 0.62, 0.46, 15.78, 0.49, 0.61) Recipe 6

8. Recipe Competition Between Active Particles Competition between active particles 97 * (226.76, 3.11, 9.61, 0.15, 0.88, 43.35, 0.44, 1.0) 38 * (57.47, 9.99, 35.18, 0.15, 0.37, 30.96, 0.05, 0.31) 56 * (15.25, 13.58, 3.82, 0.3, 0.8, 39.51, 0.43, 0.65) 31 * (113.21, 18.25, 38.21, 0.62, 0.46, 15.78, 0.49, 0.61) 67 * (216.35, 11.75, 7.7, 0.83, 0.97, 97.31, 0.02, 0.38) 29 * (254.64, 7.28, 7.0, 0.95, 0.11, 22.41, 0.43, 0.31) 13 * (105.4, 3.55, 5.24, 0.34, 0.18, 23.53, 0.39, 0.24) competition function winner: 8

9. 97 * (226.76, 3.11, 9.61, 0.15, 0.88, 43.35, 0.44, 1.0) 38 * (57.47, 9.99, 35.18, 0.15, 0.37, 30.96, 0.05, 0.31) 56 * (15.25, 13.58, 3.82, 0.3, 0.8, 39.51, 0.43, 0.65) 31 * (113.21, 18.25, 38.21, 0.62, 0.46, 15.78, 0.49, 0.61) Mutations Recipe transmission & mutation 75 * (216.35, 11.75, 7.7, 0.83, 0.97, 97.31, 0.02, 0.38) 29 * (254.64, 7.28, 7.0, 0.95, 0.11, 28.56, 0.43, 0.31) 13 * (105.4, 3.55, 5.24, 0.34, 0.18, 23.53, 0.39, 0.24) 9 67 * (216.35, 11.75, 7.7, 0.83, 0.97, 97.31, 0.02, 0.38) 29 * (254.64, 7.28, 7.0, 0.95, 0.11, 22.41, 0.43, 0.31) 13 * (105.4, 3.55, 5.24, 0.34, 0.18, 23.53, 0.39, 0.24)

10. Sayama, IEEE ALIFE (2011) Sayama & Wong, ECAL (2011) Is This Really Creative?

11. Simulate It Much Longer Detect Objects Automatically

12. Automated Object Harvesting

14. Environmental Perturbations • Competition function switched from “majority-relative” to either “faster” or “slower” for T steps in every 2000 steps in either half of the space T = 50 (short, moderate) T = 500 (long, severe)

15. Detailed Object Tracking (Object Size) With moderate environmental perturbations Objects tend to get bigger over time

16. Detailed Object Tracking (Object Size) With moderate environmental perturbations With severe environmental perturbations But it also depends on the severity of perturbations

17. Detailed Object Tracking (Color Diversity) With moderate environmental perturbations With severe environmental perturbations Severe perturbations may increase color entropy of objects

18. Results https://youtu.be/BpN-DE3o6u8

19. Evolved Under Moderate Perturbations

20. Evolved Under Severe Perturbations

21. Conclusions Evolutionary swarm chemistry can produce patterns with no global fitness evaluation The “bottom-up” approach offers a unique path to machine creativity Automated object harvesting has captured a large number of patterns

22. Thank You @hirokisayama

Artificial Creativity of Evolutionary Swarm Systems

More Related Content

More from Hiroki Sayama (13)

Recently uploaded (20)

Artificial Creativity of Evolutionary Swarm Systems