Stanley, A. E., Ashlock, D. & Testatsion, L. (1994) Iterated Prisoner's Dilemma with Choice and Refusal of Partners. In C G Langton (ed) Artificial Life III, Addison-Wesley, Redwood Cit y, pp 131-146.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....on the second factor, the biasing of interactions between in dividuals. There are several mechanisms which can generate non random interactions between individuals. For instance, individuals can bias their interactions based on past experience with and recogni tion of particular individuals ([Stanley ctal. 1994], Crowley et al. 1996] or individuals can be located in some spatial topology (e.g. i or 2 dimensional grids) with a bias to interact with nearby individuals ( Axelrod, 1984] Grim, 1996] Hoffmann and Waring, 1996] This paper explores a third way to bias interactions: the use of ....

Stanley, E. Ann, Ashlock, Dan and Tesfatsion, Leigh. "Iterated Prisoner's Dilemma with Choice and Refusal of Partner's." Artificial Life IV, pp131-176. Christopher G. Langton (ed.) Addison-Wesley, Reading MA.

....on the second factor, the biasing of interactions between individuals. There are several mechanisms which can be used to generate nonrandom interactions between individuals, including: Individuals can bias their interactions based on past experience with and recognition of particular individuals ([Stanley et al., 1994], Crowley et al. 1996] Individuals can be located in some spatial topology with a bias to inter act with nearby individuals, e.g. one dimensional grids ( Oliphant, 1994] Hoffmann and Waring, 1996] or two dimensional grids (pp.158 168 of [Axelrod, 1984] Nowak and May, 1992] Grim, ....

Stanley, E. Ann, Ashlock, Dan and Tesfatsion, Leigh. "It- erated Prisoner's Dilemma with Choice and Refusal of Partner's." Artificial Life IV, pp131-176. Christopher G. Langton (ed.) Addison-Wesley, Read- ing MA. 34

....and their evolutionary dynamics, and found that usually in the long run with no noise, there were some evolutonary stable strategies. Some other researchers investigated IPD in different contexts: with automata as decision engines, instead of rule tables [7] with choice and refusal of partners[3], or even in a spatial setting [8] but so far, the open ended (or infinite memory) evolution of IPD strategies had not been studied. Due to hardware limits it is of course impossible to have really infinite memories for each prisoner, but it can be considered infinite if the memory limits are ....

L. Tesfatsion E. A. Stanley, D. Ashlock. Iterated Prisoner's Dilemma with choice and refusal of partners. In C. G. Langton, editor, Artificial Life III. Addison-Wesley, Reading, MA, 1993.

....to explore behavior in the iterated Prisoner s Dilemma. Under tagging, agents are able to #recognize one another via an observable marker, and based on this observation decide whether or not to interact. Holland #1993# suggested that tags might allow new patterns of social interaction to develop. Stanley et al. #1994# allowed partners to recognize one another and to base refusals to play in a Prisoner s Dilemma on past experience. In Riolo #1996#, agents were more likely to interact with other agents that looked alike #based on a prede#ned metric#. Lindgren and Nordahl #1994# explored a more extended tag ....

Stanley, E. Ann, Dan Ashlock, and Leigh Tesfatsion, #Iterated Prisoner's Dilemma with Choice and Refusal of Partner's," in Arti#cial Life III, Christopher G. Langton #ed.#, Addison-Wesley: Reading, MA, 1994.

....stochastic ones ( 11] the Pavlov strategy ( 13] that cooperates when it has played the same move as its opponent etc. In the literature we may also find studies in an evolutionary perspective ( 6] theoretical or applied biological studies ( 3] 5] 10] and studies of modified IPD versions ([14]) We have designed an adaptive tit for tat strategy that is analyzed in detailed elsewhere ( 15] and is shown to score better than pure tit for tat and most of the other known strategies. Our initial motivation for this work was to find a strategy that would be able to cooperate with ....

Stanley, E.A., Ashlock, D., and Tesfatsion, L.: Iterated Prisoner's Dilemma with Choice and Refusal of Partners, Artificial Life III, Addison-Wesley (1994)

....that cooperates when it has played the same move as its opponent etc. In the literature we may also find studies in an evolutionary perspective (Fogel 1993) theoretical or applied biological studies (Axelrod Dion 1988, Feldman Thomas 1987, Milinski 1987) and studies of modified IPD versions (Stanley et al. 1994). The best designed behavior found so far in the literature is GRADUAL (Beaufils et al. 1996) which manages to achieve the highest scores against virtually all other designed behaviors. This behavior starts by cooperating and then plays Tit For Tat, except that it does not defect just once to an ....

Stanley, E.A., Ashlock, D., and Tesfatsion, L. (1994) Iterated Prisoner's Dilemma with Choice and Refusal of Partners. Artificial Life III, Addison-Wesley.

....lack of such research. While there is a small core of work on the iterated prisoner s dilemma, a classic game theory problem in which the emergence of cooperative behavior among nonaltruistic agents can be explored, and which is of direct relevance to oligopolistic markets (see, e.g. 4] [55], and [43] we know of only two papers published in the a life literature that explicitly study market trading strategies: 47] and [24] Both of these papers report on the application of elementary evolutionary adaptation methods to optimize simple trading strategies for speculative markets, and ....

E. A. Stanley, D. Ashlock, & L. Tesfatsion. Iterated prisoner's dilemmawith choice and refusal of partners. In C. G. Langton, ed, Artificial Life III, pp.131--176, Addison-Wesley, 1993.

....other on each iteration. The principle of preferential partner selection removes this constraint by allowing individuals to have some control over who they interact with, and this extension can lead to the emergence of interesting new dimensions of emergent behavioural structure. Stanley et al. [4], 3] have published extensively on the formation of social networks in an IPD context where agents can choose who they would prefer to interact with, and refuse overtures from partners they consider unsuitable. Choice and refusal is accomplished with reference to continuously updated expected ....

....for any possible game history of up to 6 prior interactions) and the actual history of interactions (if any) between the particular server customer pairing that is being considered. 3 Implementation The implementation of the model here bears certain similarities to the IPD CR model of Stanley [4], in that customers choose servers on the basis of the expected payoff values that they maintain 5 . But whereas in the standard implementation of IPD CR the agents have to play as many rounds of IPD as they have tolerable offers pending, in the call admission model the total number of analysis ....

Stanley E.A., Ashlock D., and Smucker M.D. Iterated prisoner's dilemma with choice and refusal of partners: Evolutionary results. In Moran F., Moreno A., Merelo J.J., and Chacon P., editors, Advances in Artificial Life : Lecture Notes in Artificial Intelligence. Springer-Verlag, 1995.

....other hand, does explore issues related to both cooperation and competition, but its primary focus is on the emergence of intelligent behavior in a population of agents. For example, work on the iterated prisoner s dilemma has been useful to characterize mating habits (Sandholm and Crites 1995; Stanley, Ashlock, and Tesfatsion 1993). Recently, work has begun to appear that focuses on learning in multi agent systems (Grefenstette 1991; Schmidhuber 1996; Tan 1993) Problems in multi agent systems are distinct from problems in DAI and distributed computing, from which the field was derived, in that DAI and distributed computing ....

.... of TBCL) can be extended to alternating Markov games (in which players take turns) Littman 1996) team games (in which teams of players cooperate to devise mutual strategies) Tambe 1996a; Tambe 1996b) and community games (in which players choose opponents to maximize their personal payoff) (Stanley, Ashlock, and Tesfatsion 1993). They can also be applied to games that are more traditional with homogeneous agents such as backgammon, checkers, and othello. The strengths of the approach include 30 the relative simplicity in storing examples and updating value estimates for game play. Unfortunately, the approach is both ....

Stanley, E., Ashlock, D., and Tesfatsion, L. (1993). Iterated prisoner's dilemma with choice and refusal of partners. In Proceedings of Alife III. Sante Fe Institute.

.... 1993, Nowak et al., 1994a,b, Lindgren and Nordahl, 1994, Killingback and Doebeli, 1996, and for a general introduction to cellular automata in this context, Durrett and Levin, 1994) Again, if players have the possibility of choice and refusal of partners, cooperation becomes much easier to achieve (Stanley, Ashlock and Tesfatsion, 1994). In both cases, aggregation of cooperators occurs through local rules. In contrast, cooperation becomes much more unlikely if more than two players engage in the interaction. For extensive simulations of this setup we refer to Hauert and Schuster (1997) It should be stressed that the PD is ....

Stanley, E.A., Ashlock, D. and Tesfatsion, L. (1994) Iterated Prisoner's Dilemma with choice and refusal of partners, in: Artificial Life III, ed. C.G. Langton, Addison-Wesley, Redwood City.

....to explore behavior in the iterated Prisoner s Dilemma. Under tagging, agents are able to recognize one another via an observable marker, and based on this observation decide whether or not to interact. Holland (1993) suggested that tags might allow new patterns of social interaction to develop. Stanley et al. 1994) allowed partners to recognize one another and to base refusals to play in a Prisoner s Dilemma on past experience. In Riolo (1996) agents were more likely to interact with other agents that looked alike (based on a predefined metric) Lindgren and Nordahl (1994) explored a more extended tag ....

Stanley, E. Ann, Dan Ashlock, and Leigh Tesfatsion, "Iterated Prisoner's Dilemma with Choice and Refusal of Partner's," in Artificial Life III, Christopher G. Langton (ed.), Addison-Wesley: Reading, MA, 1994.

....at the level of individual organisms and their interactions. There is a large number of models of this kind, many of them using various game theoretic approaches for the interaction between individuals, see, e.g. Matsuo 1985, Axelrod 1987, Miller 1989, Lindgren 1992, Nowak and Sigmund 1993, Stanley et al. 1993, Lindgren and Nordahl 1994a] 4 Figure 1. Evolutionary models characterized by level and dimensionality. The numbers in the figure refer to various models: 1) Stenseth and Maynard Smith 1984] 2) Kauffman 1993] 3) Anderson 1983, Stein and Anderson 1984] 4) Matsuo 1985, Axelrod 1987, ....

....models characterized by level and dimensionality. The numbers in the figure refer to various models: 1) Stenseth and Maynard Smith 1984] 2) Kauffman 1993] 3) Anderson 1983, Stein and Anderson 1984] 4) Matsuo 1985, Axelrod 1987, Miller 1989, Lindgren 1992, Nowak and Sigmund 1993, Stanley et al. 1993, Lindgren and Nordahl 1994a] 5) Holland 1992, 1993, Langton 1995] 6) Rasmussen et al. 1992, Ray 1992] 7) Eigen 1971, Schuster 1986, Fontana 1992, Kauffman 1986, Farmer et al. 1986, Eigen et al. 1988, Bagley et al. 1992] 8) Boerlijst and Hogeweg 1991, 1992] see also [Eigen and Schuster ....

[Article contains additional citation context not shown here]

Stanley, E. A., Ashlock, D., and Tesfatsion, L. (1993). "Iterated Prisoner's Dilemma with choice and refusal of partners," pp. 131-175 in C. G. Langton (Ed.) Artificial Life III (Addison-Wesley, Reading, MA).

....set at 0.005 (per bit for point mutations) 7 For clarity, the Kolmogorov complexity metric is omitted; it followed the same pattern as the memory metric. 8 A further change in this experiment was the introduction of partner choice mechanisms for each individual, based on an algorithm from Stanley, Ashlock, Smucker (1995), and discussed in detail in Seth (1997) The effect of this mechanism was to further facilitate the evolution of complex strategies, even to the extent that a small fitness cost on long genotypes could be overcome. Figure 2: Evolution of complexity without variability; complexity does not evolve ....

Stanley, E.; Ashlock, D.; and Smucker, M. 1995. Iterated prisoner's dilemma with choice and refusal of partners: Evolutionary results. In Moran, F.; Moreno, A.; Merelo, J.; and Chacon, P., eds., Advances in Artificial Life : Lecture Notes in Artificial Intelligence. Springer-Verlag.

....we normally exercise considerable discretion about who we interact with, rather than being forced to interact with everyone. This principle of partner selection allows new dimensions of emergent behavioural structure to develop, and has been introduced into a number of IPD models. Stanley et al. [19], 2] 18] have published a series of papers looking at the formation of social networks in an IPD context with choice and refusal permitted. Agents choose and refuse with reference to continuously updated expected payoffs that each agent maintains for every other agent in the population (they ....

....of the population maintains an expectation value for every other member, reflecting the expected outcome of a round of prisoner s dilemma. This value, set initially to 3. 0 (mutual cooperation) is allowed to alter on the basis of experience, as illustrated by the pseudocode overleaf, based on [19]) FOR EACH population member sort remainder on basis of expectation select the most preferable (up to a quota) and make offers; recipients thus form a list of offers ENDFOREACH FOR EACH population member FOR EACH offer received IF the offer is tolerable play one iteration and update scores and ....

[Article contains additional citation context not shown here]

E.A. Stanley, D. Ashlock D., and M.D. Smucker. Iterated prisoner's dilemma with choice and refusal of partners: Evolutionary results. In F. Moran, A. Moreno, J.J. Merelo, and P. Chacon, editors, Advances in Artificial Life : Lecture Notes in Artificial Intelligence. SpringerVerlag, 1995.

....model, in which every member was forced to interact with every other member, and a choice and refusal model (the IPD CR) in which each member maintained a set of expected payoff values for every other member, and could choose and refuse who to interact with. This latter model is drawn from [14] and is redescribed in [9] A tournament style genetic algorithm was employed in both models, with equal probabilities for genotype doubling, splitting, and point mutations (see Appendix 1 for parameter values) 2.2 The IPD and the LRV Recalling the description of Ashby s LRV in Section 1.1, we ....

E.A. Stanley, D. Ashlock, and M.D. Smucker. Iterated prisoner's dilemma with choice and refusal of partners: Evolutionary results. In F. Moran, A. Moreno, J.J. Merelo, and P. Chacon, editors, Advances in Artificial Life : Lecture Notes in Artificial Intelligence. Springer-Verlag, 1995.

....EMPLOYER 1 Introduction An evolutionary match and play game is an evolutionary game in which the concept of rational play is extended to include the choice and refusal of partners as well as the choice of strategy to play with any given partner. Such games have previously been studied by Stanley et al. #1994#, Ashlock et al. #1996#, Hauk #1996#, and Tesfatsion #1997a#. Other game theory studies that have allowed players to avoid unwanted interactions, or more generally to a#ect the probabilityofinteraction with other players through their own actions, include Fogel #1995#, Hirshleifer and Rasmusen ....

....characteristics and the formation and evolution of trade networks, and between trade network formation and the types of trade behavior and social welfare outcomes that these trade networks support. These descriptive statistics complement and extend the descriptive statistics developed by Stanley et al. #1994#, Smucker et al. #1994#, and Ashlocket al... #1996# to characterize play behavior and signi#cant play graphs. The main conclusion drawn from this study is that the optimality criteria conventionally used to evaluate the performance of matching mechanisms in static market contexts turn out to be ....

Stanley, E. A., Ashlock, D. and Tesfatsion, L. #1994#. #Iterated Prisoner's Dilemma with Choice and Refusal of Partners," in Arti#cial Life III #C.

....studies are set in static contexts [18] Since actual social interactions are often characterized both by evolutionary dynamics and by preference based partner selection, studying both aspects together seems a logical and interesting next step to take. This issue is addressed in Stanley et al. [20]. The standard evolutionary iterated prisoner s dilemma (IPD) is extended to an evolutionary IPD with choice and refusal (IPD CR) by allowing players to choose and refuse game partners in each iteration on the basis of continually updated expected payoffs. 2 The introduction of choice and ....

....of defectors occurs endogenously as an increasing number of players individually refuse the defectors game offers. Nevertheless, choice and refusal also permit opportunistic players to home in quickly on exploitable players and form parasitic relationships. The computer experiments reported in [20] and in the subsequent studies by Ashlock et al. 1] Smucker et al. 19] and Hauk [8] indicate that the emergence of mutual cooperation in the standard evolutionary IPD is accelerated by the introduction of choice and refusal of partners. The underlying player interaction patterns induced by ....

[Article contains additional citation context not shown here]

Stanley, E. A., Ashlock, D., and Tesfatsion, L., Iterated prisoner's dilemma with choice and refusal of partners, in C. Langton (ed.), Artificial Life III, Proceedings Volume 17, Santa Fe Institute Studies in the Sciences of Complexity, Addison Wesley, 1994, pp. 131--175.

....C8, C9, D8, D4. 1 Introduction An evolutionary match and play game is an evolutionary game in which the concept of rational play is extended to include the choice and refusal of partners as well as the choice of strategy to play with any given partner. Such games have previously been studied by Stanley et al. 1994), Ashlock et al. 1996) Hauk (1996) and Tesfatsion (1997a) Other game theory studies that have allowed players to avoid unwanted interactions, or more generally to affect the probability of interaction with other players through their own actions, include Fogel (1995) Hirshleifer and Rasmusen ....

....characteristics and the formation and evolution of trade networks, and between trade network formation and the types of trade behavior and social welfare outcomes that these trade networks support. These descriptive statistics complement and extend the descriptive statistics developed by Stanley et al. 1994), Smucker et al. 1994) and Ashlock et al. 1996) to characterize play behavior and significant play graphs. The main conclusion drawn from this study is that the optimality criteria conventionally used to evaluate the performance of matching mechanisms in static market contexts turn out to be ....

Stanley, E. A., Ashlock, D. and Tesfatsion, L. (1994). "Iterated Prisoner's Dilemma with Choice and Refusal of Partners," in Artificial Life III (C.

No context found.

Stanley, E. A., D. Ashlock, and L. Tesfatsion (1994), "Iterated Prisoner's Dilemma with Choice and Refusal of Partners," 131-175 in C. Langton, ed., Artificial Life III , Proceedings Volume 17, SFI Studies in the Sciences of Complexity, Addison-Wesley, Reading, MA.

....of a decentralized market economy, the labor market framework demonstrates how an ACE approach facilitates the modelling of markets from an agent based perspective and permits the rigorous experimental study of non steady state dynamics. The framework builds on a series of earlier studies [7, 56, 57, 59, 60, 61]. 5 The ACE labor market framework is implemented by means of the C trade network game (TNG) framework (version 105b) developed by McFadzean and Tesfatsion [41] which in turn is supported by SimBioSys, a general C class library for evolutionary simulations developed by McFadzean [40] The ....

Stanley, E. A., Ashlock, D., and Tesfatsion, L., "Iterated Prisoner's Dilemma with Choice and Refusal of Partners," pp. 131--175 in Langton, C. (ed.), Artificial Life III , Proceedings Volume 17, Santa Fe Institute Studies in the Sciences of Complexity, Addison Wesley, 1994.

No context found.

Stanley, E. A., D. Ashlock, and L. Tesfatsion (1994), "Iterated Prisoner's Dilemma with Choice and Refusal of Partners," 131-175 in C. Langton, ed., Artificial Life III , Proceedings Volume 17, SFI Studies in the Sciences of Complexity, Addison-Wesley, Reading, MA.

....it is better for players to pay for information about a stranger or to costlessly assume that they are like themselves. Here we assume that all strangers look the same, and focus more on the question of what individuals do once they have some information about another player. In a previous paper (Stanley et al. 1994), we studied an IPD choice and refusal mechanism that combines active choice of potential game partners with the ability to refuse play with those judged to be intolerable. Players use continually updated expected payoffs to assess the relative desirability of potential partners. This use of ....

....as the shame of being refused whereas being approached is costless. In addition, a (possibly small) cost is imposed on wallflower players who neither make nor accept offers, which tends to deter this kind of behavior. We study a slightly simplified version of the choice and refusal tournament of Stanley et al. 1994). Each generation plays a tournament consisting of I iterations. At the beginning of each iteration i 1, each player n associates an expected payoff with each other player m, denoted by i Gamma1 (mjn) This expected payoff is used to determine which players are tolerable as partners and which ....

[Article contains additional citation context not shown here]

E. A. Stanley, D. Ashlock, & L. Tesfatsion (1994) "Iterated Prisoner's Dilemma with Choice and Refusal of Partners," In: Artificial Life III, C. G. Langton (ed.), SFI Studies in the Sciences of Complexity, Proc. Vol. 17, Addison-Wesley, pp. 131--176.

....network of agent interactions. The present paper focuses on the particular ACE model developed by Tesfatsion [15] to study the endogenous formation and evolution of trade networks. This model, referred to as the trade network game (TNG) extends to an economic setting an earlier model ( 1] [13]) combining evolutionary game play with endogenous partner selection. In the TNG, successive generations of resource constrained traders choose and refuse trade partners on the basis of continuously updated expected payoffs, engage in risky trades modelled as twoperson games, and evolve their ....

.... bioWorld to situate the tradebots in a virtual spatial environment subject both to biological processes (e.g. plant growth) and to physical laws (e.g. conservation of energy) 7 The experimental discovery of Rip Off was made by Daniel Ashlock of Iowa State University during the preparation of [13]. 8 Technically, however, any tradebot instance can access the private member functions and data of other tradebots within the scope of a tngTradeBot method. class tngTradeBot f Public Access: Internalized Institutional Rules Methods for determining my trade partners; Methods for ....

Stanley, E. A., Ashlock, D., and Tesfatsion, L., Iterated prisoner's dilemma with choice and refusal of partners, pp. 131--175 in C. Langton (ed.), Artificial Life III, Proceedings Volume 17, Santa Fe Institute Studies in the Sciences of Complexity, Addison Wesley, 1994.

....Tesfatsion. The present paper discusses the C implementation of a particular ACE model developed by Tesfatsion [7] to study the endogenous formation and evolution of trade networks. This model, referred to as the trade network game (TNG) extends to an economic setting an earlier model ( 1] [6]) combining evolutionary game play with endogenous partner selection. In the TNG, successive generations of resource constrained traders choose and refuse trade partners on the basis of continuously updated expected payoffs, engage in risky trades modelled as two person games, and evolve their ....

E. A. Stanley, D. Ashlock, and L. Tesfatsion. Iterated prisoner's dilemma with choice and refusal of partners. pp. 131--175 in C. Langton (ed.), Artificial Life III. Proceedings Volume 17. Santa Fe Institute Studies in the Sciences of Complexity, Addison Wesley, 1994.

No context found.

Stanley, A. E., Ashlock, D. & Testatsion, L. (1994) Iterated Prisoner's Dilemma with Choice and Refusal of Partners. In C G Langton (ed) Artificial Life III, Addison-Wesley, Redwood Cit y, pp 131-146.

No context found.

E.A. Stanley, D. Ashlock, and L. Tesfatsion, "Iterated prisoner's dilemma with choice and refusal of partners," Artificial Life III, C.G. Langton (ed.), Addison-Wesley, Reading, MA, pp. 131-175, 1994.

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