Nolfi, Stefano and Floreano, Dario (1998), Coevolving Predator and Prey Robots: Do "Arms Races" rise in Artificial Evolution? In Artificial Life 4:311-335.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....2. #right ##left) 3. 3 Multi agent Systems: The Pursuit Game Example tor Pre (or Pursuit) wide riet of a proa hes nd it s ma di#erent insta tia ions thaca be used e di#erent multi afi t scenaSyI [25] As the Zigza:fiyI exa:fiy nces of the tor Pre ve been modeled utonomous robots [16]. Here we modela simple nce of this me. tors a d pre moveaL nd discrete, grid like toroida l world withsqua respa ces; n move o# one end of the boa rda comeba ck on the other end. Preda torsa nd pre move simulta neousl . n move vertica horizonta n direction. In order to simula st but not ver ....

S. Nolfi and D. Floreano. Coevolving Predator and Prey Robots: Do "Arms Races" Arise in Artificial Evolution? Artificial Life, 4(4):311--335, 1998.

....still has not been solved. The predator prey pursuit problem was introduced by Benda et al. 3] and comprised four predator agents whose goal are to capture a prey agent by surrounding it on four sides in a grid world. This problem has been used to study phenomena such as competitive co evolution [16][18] 10] multi agent strategies, and multi agent communication. The rest of this paragraph describes some previous studies on the latter two phenomena that made use of the predator prey pursuit problem. Haynes and Sen [10] used genetic programming to evolve predator strategies and showed that a ....

Stefano Nolfi and Dario Floreano. Coevolving predator and prey robots: Do "arms races" arise in artificial evolution? Artificial Life, 4(4):337--357, 1998.

....of competitive co evolution. For example, in a companion paper we report results obtained in different environmental conditions where predators evolved against a fixed co evolved prey (from another run) do not reach the same performance levels obtained when both prey and predators are co evolved [29]. In other words, under certain circumstances competitive co evolution can indeed produce more powerful solutions that standard evolution. Furthermore, it has also been shown that, by including all the best opponents evolved so far as test cases for each individual (Hall of Fame method) ....

....solutions that standard evolution. Furthermore, it has also been shown that, by including all the best opponents evolved so far as test cases for each individual (Hall of Fame method) co evolution displays monotonic progress and becomes very similar to a very robust optimization technique [31, 29]. Finally, other results obtained on variations of the basic physical setup described in this article display ascending trends on the Master Tournaments [29] However, the basic issue here is to what extent natural evolution should be interpreted as an optimization process (see also [17, 16] for a ....

[Article contains additional citation context not shown here]

S. Nolfi and D. Floreano. Co-evolving predator and prey robots: Do "arms races" arise in artificial evolution? Artificial Life, 1998. To be published.

....the real robots. hint at the computational advantages of competitive coevolution. In a companion paper, we report results showing that predators evolved against a fixed co evolved prey (from another run) do not reach the same performance levels obtained when both prey and predators are co evolved (Nolfi and Floreano, 1998). It has also been shown that, by including all the best opponents evolved so far as test cases for each individual (Hall of Fame method) co evolution becomes very similar to a very robust optimization technique (Rosin and Belew, 1997) However, the basic issue that the results presented above ....

....To this extent, co evolution of predator prey scenarios, such as the one described in this paper, not only displays these properties, but also provides a framework for understanding some of the issues brought in by open ended evolution and testing different methodologies. Indeed, elsewhere (Nolfi and Floreano, 1998) we have shown that under certain conditions plain competitive co evolution can provide better adaptativity than single agent evolution and methods aimed at improving the (traditional) optimization properties of co evolution, such as the Hall of Fame mentioned above. The results described above ....

Nolfi, S. and Floreano, D. (1998). Co-evolving predator and prey robots: Do "arms races" arise in artificial evolution? Artificial Life. To be published.

No context found.

Nolfi, Stefano and Floreano, Dario (1998), Coevolving Predator and Prey Robots: Do "Arms Races" rise in Artificial Evolution? In Artificial Life 4:311-335.

No context found.

S. Nolfi and D. Floreano. Coevolving predator and prey robots: Do "arms races" arise in artificial evolution? Artificial Life, 4(4):311--335, 1998.

No context found.

Stefano Nolfi and Dario Floreano. Coevolving predator and prey robots: Do "arms races" arise in artificial evolution? Artificial Life, 4(4):311--335, 1998.

No context found.

S. Nolfi and D. Floreano. Coevolving predator and prey robots: Do "arms races" arise in artificial evolution? Artificial Life, 4(4):311-- 335, 1998. 34

No context found.

S. Nolfi and D. Floreano. Coevolving predator and prey robots: Do "arms races" arise in artificial evolution? Artificial Life, 4(4):311--335, 1998.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC