Evolving optimum camouflage with Generative Adversarial Networks
- Resource Type
- Authors
- Innes C. Cuthill; John G. Fennell; Laszlo Talas; Nicholas E. Scott-Samuel; Karin Kjernsmo; Roland J. Baddeley
- Source
- Subject
- 0106 biological sciences
0303 health sciences
Exploit
business.industry
Computer science
010603 evolutionary biology
01 natural sciences
GeneralLiterature_MISCELLANEOUS
Predation
03 medical and health sciences
Adversarial system
Evolutionary arms race
Camouflage
Artificial intelligence
business
Generative grammar
030304 developmental biology
- Language
We describe a novel method to exploit Generative Adversarial Networks to simulate an evolutionary arms race between the camouflage of a synthetic prey and its predator. Patterns evolved using our methods are shown to provide progressively more effective concealment and outperform two recognised camouflage techniques. The method will be invaluable, particularly for biologists, for rapidly developing and testing optimal camouflage or signalling patterns in multiple environments.