Don't Move: The T-Rex Effect in the Predator-Prey World

Todd M. Bacastow, Kam Chuen Jim, Brian R. Cook, C. Lee Giles

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

To develop robust agents capable of adapting to changes in environment conditions we analyze the effects of introducing noise into multi-agent communication and varying prey movement patterns in a version of the predator-prey pursuit problem. In this simulation, predators communicate with languages evolved using a genetic algorithm. We show the time it takes to capture prey increases as more noise is introduced into the predators' communication string. We show that predator performance under noisy conditions can be improved by training the predators with noise. Increasing the amount of noise during predator training increases the performance capability of the predator, thus reducing the amount of time it takes to capture a prey in noisy conditions. Training against different prey movement patterns can also enhance predator performance. We show that predators trained with prey that stop and move randomly perform best when tested against various prey movements. Alternatively, predators trained against constantly idle, or frozen, prey showed the worst performance. When any type of predator was tested against the frozen prey, the predators had some problems completing the capture. Drawing inspiration from the popular film Jurassic Park, we refer to this phenomenon as the T-Rex effect. Prey that hide or do not move have the greatest probability of surviving. Training predators with factors that normally decrease testing performance helps develop robust predators capable of increased performance under both adverse testing conditions.

Original languageEnglish (US)
Title of host publicationProceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03
EditorsJ.S. Rosenschein, T. Sandholm, M. Wooldridge, M. Yakoo
Pages924-925
Number of pages2
Volume2
StatePublished - 2003
EventProceedings of the Second International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS 03 - Melbourne, Vic., Australia
Duration: Jul 14 2003Jul 18 2003

Other

OtherProceedings of the Second International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS 03
CountryAustralia
CityMelbourne, Vic.
Period7/14/037/18/03

Fingerprint

Communication
Testing
Genetic algorithms

All Science Journal Classification (ASJC) codes

  • Engineering(all)

Cite this

Bacastow, T. M., Jim, K. C., Cook, B. R., & Lee Giles, C. (2003). Don't Move: The T-Rex Effect in the Predator-Prey World. In J. S. Rosenschein, T. Sandholm, M. Wooldridge, & M. Yakoo (Eds.), Proceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03 (Vol. 2, pp. 924-925)
Bacastow, Todd M. ; Jim, Kam Chuen ; Cook, Brian R. ; Lee Giles, C. / Don't Move : The T-Rex Effect in the Predator-Prey World. Proceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03. editor / J.S. Rosenschein ; T. Sandholm ; M. Wooldridge ; M. Yakoo. Vol. 2 2003. pp. 924-925
@inproceedings{8390d6385eda48eea6d09b70de11de0d,
title = "Don't Move: The T-Rex Effect in the Predator-Prey World",
abstract = "To develop robust agents capable of adapting to changes in environment conditions we analyze the effects of introducing noise into multi-agent communication and varying prey movement patterns in a version of the predator-prey pursuit problem. In this simulation, predators communicate with languages evolved using a genetic algorithm. We show the time it takes to capture prey increases as more noise is introduced into the predators' communication string. We show that predator performance under noisy conditions can be improved by training the predators with noise. Increasing the amount of noise during predator training increases the performance capability of the predator, thus reducing the amount of time it takes to capture a prey in noisy conditions. Training against different prey movement patterns can also enhance predator performance. We show that predators trained with prey that stop and move randomly perform best when tested against various prey movements. Alternatively, predators trained against constantly idle, or frozen, prey showed the worst performance. When any type of predator was tested against the frozen prey, the predators had some problems completing the capture. Drawing inspiration from the popular film Jurassic Park, we refer to this phenomenon as the T-Rex effect. Prey that hide or do not move have the greatest probability of surviving. Training predators with factors that normally decrease testing performance helps develop robust predators capable of increased performance under both adverse testing conditions.",
author = "Bacastow, {Todd M.} and Jim, {Kam Chuen} and Cook, {Brian R.} and {Lee Giles}, C.",
year = "2003",
language = "English (US)",
volume = "2",
pages = "924--925",
editor = "J.S. Rosenschein and T. Sandholm and M. Wooldridge and M. Yakoo",
booktitle = "Proceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03",

}

Bacastow, TM, Jim, KC, Cook, BR & Lee Giles, C 2003, Don't Move: The T-Rex Effect in the Predator-Prey World. in JS Rosenschein, T Sandholm, M Wooldridge & M Yakoo (eds), Proceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03. vol. 2, pp. 924-925, Proceedings of the Second International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS 03, Melbourne, Vic., Australia, 7/14/03.

Don't Move : The T-Rex Effect in the Predator-Prey World. / Bacastow, Todd M.; Jim, Kam Chuen; Cook, Brian R.; Lee Giles, C.

Proceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03. ed. / J.S. Rosenschein; T. Sandholm; M. Wooldridge; M. Yakoo. Vol. 2 2003. p. 924-925.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Don't Move

T2 - The T-Rex Effect in the Predator-Prey World

AU - Bacastow, Todd M.

AU - Jim, Kam Chuen

AU - Cook, Brian R.

AU - Lee Giles, C.

PY - 2003

Y1 - 2003

N2 - To develop robust agents capable of adapting to changes in environment conditions we analyze the effects of introducing noise into multi-agent communication and varying prey movement patterns in a version of the predator-prey pursuit problem. In this simulation, predators communicate with languages evolved using a genetic algorithm. We show the time it takes to capture prey increases as more noise is introduced into the predators' communication string. We show that predator performance under noisy conditions can be improved by training the predators with noise. Increasing the amount of noise during predator training increases the performance capability of the predator, thus reducing the amount of time it takes to capture a prey in noisy conditions. Training against different prey movement patterns can also enhance predator performance. We show that predators trained with prey that stop and move randomly perform best when tested against various prey movements. Alternatively, predators trained against constantly idle, or frozen, prey showed the worst performance. When any type of predator was tested against the frozen prey, the predators had some problems completing the capture. Drawing inspiration from the popular film Jurassic Park, we refer to this phenomenon as the T-Rex effect. Prey that hide or do not move have the greatest probability of surviving. Training predators with factors that normally decrease testing performance helps develop robust predators capable of increased performance under both adverse testing conditions.

AB - To develop robust agents capable of adapting to changes in environment conditions we analyze the effects of introducing noise into multi-agent communication and varying prey movement patterns in a version of the predator-prey pursuit problem. In this simulation, predators communicate with languages evolved using a genetic algorithm. We show the time it takes to capture prey increases as more noise is introduced into the predators' communication string. We show that predator performance under noisy conditions can be improved by training the predators with noise. Increasing the amount of noise during predator training increases the performance capability of the predator, thus reducing the amount of time it takes to capture a prey in noisy conditions. Training against different prey movement patterns can also enhance predator performance. We show that predators trained with prey that stop and move randomly perform best when tested against various prey movements. Alternatively, predators trained against constantly idle, or frozen, prey showed the worst performance. When any type of predator was tested against the frozen prey, the predators had some problems completing the capture. Drawing inspiration from the popular film Jurassic Park, we refer to this phenomenon as the T-Rex effect. Prey that hide or do not move have the greatest probability of surviving. Training predators with factors that normally decrease testing performance helps develop robust predators capable of increased performance under both adverse testing conditions.

UR - http://www.scopus.com/inward/record.url?scp=1142304653&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=1142304653&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:1142304653

VL - 2

SP - 924

EP - 925

BT - Proceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03

A2 - Rosenschein, J.S.

A2 - Sandholm, T.

A2 - Wooldridge, M.

A2 - Yakoo, M.

ER -

Bacastow TM, Jim KC, Cook BR, Lee Giles C. Don't Move: The T-Rex Effect in the Predator-Prey World. In Rosenschein JS, Sandholm T, Wooldridge M, Yakoo M, editors, Proceedings of the Second International Joint Conference on Autonomous Agents and multiagent Systems, AAMAS 03. Vol. 2. 2003. p. 924-925