Real-time interactive learning in the NERO video game
作者： Igor Karpov , Risto Miikkulainen , Kenneth O. Stanley , Aliza Gold
DOI:
关键词:
摘要: In the NeuroEvolving Robotic Operatives (NERO) video game, player trains a team of virtual robots for combat against other players' teams. The learn in real time through interacting with player. Since NERO was originally released June, 2005, it has been downloaded over 50,000 times, appeared on Slashdot, and won several honors. real-time NeuroEvolution Augmenting Topologies (rt-NEAT) method, which can evolve increasingly complex artificial neural networks as game is being played, drives robots' learning, making possible this entirely new genre game. live demo will show how agents adapt they interact future, rtNEAT may allow kinds educational training applications interactive adapting games.
参考文章(5)
Kenneth O. Stanley, Risto Miikkulainen, Evolving neural networks through augmenting topologies Evolutionary Computation. ,vol. 10, pp. 99- 127 ,(2002) , 10.1162/106365602320169811
D.B. Fogel, T.J. Hays, D.R. Johnson, A platform for evolving characters in competitive games congress on evolutionary computation. ,vol. 2, pp. 1420- 1426 ,(2004) , 10.1109/CEC.2004.1331063
K.O. Stanley, B.D. Bryant, R. Miikkulainen, Real-time neuroevolution in the NERO video game IEEE Transactions on Evolutionary Computation. ,vol. 9, pp. 653- 668 ,(2005) , 10.1109/TEVC.2005.856210
Kenneth O. Stanley and Bobby D. Bryant and Risto Miikkulainen, Evolving Neural Network Agents in the NERO Video Game ,(2005)
K. O. Stanley, R. Miikkulainen, Competitive coevolution through evolutionary complexification Journal of Artificial Intelligence Research. ,vol. 21, pp. 63- 100 ,(2004) , 10.1613/JAIR.1338