spatial equilibrium models

Marietto, M., David, N., Sichman, J. and Coelho, H. 2003. Requirement analysis of agent-based simulation platforms: State of the art and new prospects. Lecture Notes in Artificial Intelligence:125-141.

Parker, D., Berger, T. and Manson, S., editors. 2002. Agent-based models of land-use and land-cover change. LUCC Report Series, 6, Indiana University.

Petit, O. 2001. Combining mas with gis: Another way to "pixelise" the commons? The Common Property Resource Digest. Quaterly Publication of the international association for the study of common property:9-11.

Röling, N. 1999. Modelling the soft side of the land: The potential of multi-agent systems. Pages 73-97 in C. Leeuwis, editor. Integral design: Innovation in agriculture and resource management. Mansholt Instituu te, Wageningen.

Mertens and Lambin, 2000. land cover-change trajectories in southern cameroon. annals of the association of american geographers, 93, 467-494. (um modelo economico)

D. Koller and A. Pfeffer

Work on game playing in AI has typically ignored games of imperfect information such as poker. In this paper, we present a framework for dealing with such games. We point out several important issues that arise only in the context of imperfect information games, particularly the insufficiency of a simple game tree model to represent the players’ information state and the need for randomization in the players’ optimal strategies. We describe Gala, an implemented system that provides the user with a very natural and expressive language for describing games. From a game description, Gala creates an augmented game tree with information sets which can be used by various algorithms in order to find optimal strategies for that game. In particular, Gala implements the first practical algorithm for finding optimal randomized strategies in two-player imperfect information competitive games [Koller et al., 1994]. The running time of this algorithm is polynomial in the size of the game tree, whereas previous algorithms were exponential. We present experimental results showing that this algorithm is also efficient in practice and can therefore form the basis for a game playing system.

J. H. Koeslag and E. Terblanche, 2003

‘‘Cooperation’’ defines any behavior that enhances the fitness of a group (e.g. a community or species), but which, by its nature, can be exploited by selfish individuals, meaning, firstly, that selfish individuals derive an advantage from exploitation which is greater than the average advantage that accrues to unselfish individuals. Secondly, exploitation has no intrinsic fitness value except in the presence of the ‘‘cooperative behavior’’. The mathematics is described by the simple Prisoner’s Dilemma Game (PDG). It has previously been shown that koinophilia (the avoidance of sexual mates displaying unusual or atypical phenotypic features, such as mutations) stabilizes any inherited strategy in the simple or iterated PDG, meaning that it cannot be displaced by rare formsof alternative behavior which arise through mutation or occasional migration. In the present model equal numbersof cooperatorsand defectors(in the simple PDG) were randomly spread in a two-dimensional ‘‘cornfield’’ with uniformly distributed resources. Every individual was koinophilic, and interacted (sexually and in the PDG tournaments) only with individuals from within its immediate neighborhood. This model therefore tested whether cooperation can outcompete defection or selfishness in a straight, initially equally matched, evolutionary battle. The results show that in the absence of koinophilia cooperation was rapidly driven to extinction. With koinophilia there was a very rapid loss of cooperators in the first few generations, but thereafter cooperation slowly spread, ultimately eliminating defection completely. This result was critically dependent on sampling effects of neighborhoods. Small samples (resulting from low population densities or small neighborhood sizes) increase the probability that a chance neighborhood comes to consist predominantly of cooperators. A sexual preference for the most common phenotype in the neighborhood then makes that phenotype more common still. Once this occurs cooperation’s spread becomes almost inevitable.

J. S. Banksa and J. Dugganb and M. Bretond, 2004

This paper extends the theory of the core, the uncovered set, and the related undominated set to a general set of alternatives and an arbitrary measure space of voters. We investigate the properties of social preferences generated by simple games; we extend results on generic emptiness of the core; we prove the general nonemptiness of the uncovered and undominated sets; and we prove the upper hemicontinuity of these correspondences when the voters’ preferences are such that the core is nonempty and externally stable. Finally, we give conditions under which the undominated set is lower hemicontinuous.

E. Ahmed and A. S. Hegazi and A. S. Elgazzar, 2002

The stability of some spatial asymmetric games is discussed. Both linear and nonlinear asymptotic stability of asymmetric hawk-dove and prisoner’s dilemma are studied. Telegraph reaction diffusion equa- tions for the asymmetric spatial games are presented. Asymmetric game of parental investment is studied in the presence of both ordi- nary and cross diffusions.

L. A. Bach and T. Helvik and F. B. Christiansen, 2006

An evolutionary game of individuals cooperating to obtain a collective benefit is here modelled as an n-player Prisioner's Dilemma game. With reference to biological situations, such as group foraging, we introduce a threshold condition in the number of cooperators required to obtain the collective benefit. In the simplest version, a three player game, complex behaviour appears as the replicator dynamics exhibits a catastrophic event separating a parameter region allowing for cohexistence of cooperators and defectors and a region of pure defection. Cooperation emerges through an ESS bifurcation, and cooperators only thrive beyond a critical point in defecting populations. The results illustrate the qualitative difference between two player games and multiple player games and thus the limitations to the generality of conclusions from two-player games. We present a procedure to find the ESS in any n-player game with cost and benefit depending on the number of cooperators. This was previously done by Motro in the special cases of convex and concave benefit functions and constant cost.

O. Kirchkamp, 2000

We apply the idea of evolution to a spatial model. Prisoners’ dilemmas or coordination games are played repeatedly within neighbourhoods where players do not optimise but instead copy successful strategies. Discriminative behaviour of players is introduced representing strategies as small automata, which can be in different states against different neighbours. Extensive simulations show that cooperation persists even in a stochastic environment, that players do not always coordinate on risk dominant equilibria in 2x2 coordination games, and that success among surviving strategies may differ. We present two analytical models that help understanding of these phenomena.

J. Galliard and R. Ferriere and U. Dieckmann, 2003

We study the spatial adaptive dynamics of a continuous trait that measures individual investment in altruism. Our study is based on an ecological model of a spatially heterogeneous population from which we derive an appropriate measure of fitness. The analysis of this fitness measure uncovers three different selective processes controlling the evolution of altruism: the direct physiological cost, the indirect genetic benefits of cooperative interactions, and the indirect genetic costs of competition for space. In our model, habitat structure and a continuous life cycle makes the cost of competing for space with relatives negligible. Our study yields a classification of adaptive patterns of altruism according to the shape of the costs of altruism (with decelerating, linear, or accelerating dependence on the investment in altruism). The invasion of altruism occurs readily in species with accelerating costs, but large mutations are critical for altruism to evolve in selfish species with decelerating costs. Strict selfishness is maintained by natural selection only under very restricted conditions. In species with rapidly accelerating costs, adaptation leads to an evolutionarily stable rate of investment in altruism that decreases smoothly with the level of mobility. A rather different adaptive pattern emerges in species with slowly accelerating costs: high altruism evolves at low mobility, whereas a quasiselfish state is promoted in more mobile species. The high adaptive level of altruism can be predicted solely from habitat connectedness and physiological parameters that characterize the pattern of cost. We also show that environmental changes that cause increased mobility in those highly altruistic species can beget selection-driven self-extinction, which may contribute to the rarity of social species.

H V. D. Parunak and S. A. Brueckner and R. Matthews and J. Sauter, 2006

Geospatial data are often used to predict or recommend movements of robots, people, or animals (‘walkers’). Analysis of such systems can be combinatorially explosive. Each decision that a walker makes generates a new set of possible future decisions, and the tree of possible futures grows exponentially. Complete enumeration of alternatives is out of the question. One approach that we have found promising is to instantiate a large population of simple computer agents that explore possible paths through the landscape. The aggregate behaviour of this swarm of agents estimates the likely behaviour of the real-world system. This paper will discuss techniques that we have found useful in swarming geospatial reasoning, illustrate their behaviour in specific cases, compare them with existing techniques for path planning, and discuss the application of such systems.

P. A. Graniero and V. B. Robinson, 2006

Many environmental, ecological, and social problems require investigation using a mixture of landscape models, individual-based models, and some level of interaction between them. Few simulation-modelling frameworks are structured to handle both styles of model in an integrated fashion. ECO-COSM is a framework that is capable of handling complex models with both landscape and agent components. Its Probe-based architecture allows model components to have controlled access to the state of other components. The ProbeWrapper is a modification of this common design approach which allows alterations to the state retrieved from the model and is a critical component of ECO-COSM’s broad modelling capability. It allows agents to apply perceptual filters or measurement errors to their observations of the landscape, or apply decisionmaking strategies in the face of incomplete or uncertain observations. ECOCOSM is demonstrated with a landscape model of metapopulation dynamics, an agent model of squirrel dispersal, and a coupled landscape-agent model to evaluate field-data-acquisition strategies for identifying nutrient or contaminant hotspots.

Border topology: wrapping, reflecting or hard (default?)

J. Hindriks, 1999

In a context where both the poor and the rich are (imperfectly) mobile, this paper compares the Nash equilibrium levels of income redistribution from the rich to the poor when jurisdictions compete either in taxes, in transfers or both in taxes and transfers. Although taxes and transfers are linked through the budgetbalanced requirement, the analysis reveals intriguing differences. Indeed, it turns out that transfer competition results in much less redistribution than tax competition, while taxtransfer competition involves an intermediate level of redistribution. In each approach, the mobility of the rich is detrimental to redistribution and an increase in the dependency ratio reduces taxes. Concerning the effect of the mobility of the poor, these approaches reach opposite conclusions. That is, the mobility of the poor is beneficial to redistribution under tax competition but reduces redistribution under transfer competition.

G. M. Myers and Y. Y. Papageorgiou, 1997

We consider a federation of two regions populated by identical individuals, in which interregional migration is costly. We define a federation as an economy in which migration may not be restricted by governments. We compare and contrast firstbest efficiency (with migration controls) and federal efficiency (without migration controls). We show that firstbest efficiency requires maximising total product net of migration cost, while federal efficiency does not. We also show that migration costs may lead to a discontinuous federal utilitypossibility frontier and to discontinuous regional reaction functions. We establish that decentralised equilibrium allocations may not be firstbest efficient but are federally efficient. We conclude by tying together wellunderstood results from the limiting cases of free mobility and immobility with our results for the intermediate case.

M. R. Jepsen and S. Leisz and K. Rasmussen and J. Jakobsen and L. Mollerjensen and L. Christiansen, 2006

Shifting cultivation in the Nghe An Province of Vietnam’s Northern Mountain Region produces a characteristic land-cover pattern of small and larger fields. The pattern is the result of farmers cultivating either individually or in spatially clustered groups. Using spatially explicit agent-based modelling, and relying on empirical data from fieldwork and observations for parameterization of variables, the level of clustering in agricultural fields observed around a study village is reproduced. Agents in the model act to maximize labour productivity, which is based on potential yield and labour costs associated with fencing of fields, and are faced with physical constraints. The simulation results are compared with land-cover data obtained from remote sensing. Comparisons are made on patterns as detected visually and using the mean nearest-neighbour ratio. Baseline simulation outputs show high degrees of spatial clustering and similarity to the land-cover data, but also a need for further calibration of model variables and controls.

M. Wooders and E. Cartwright and R. Selten, 2006

In the literature of psychology and economics it is frequently observed that individuals tend to conform in their behavior to that of similar individuals. A fundamental question is whether the outcome of such conformity can be consistent with self-interest. We propose that this consistency requires the existence of a Nash or approximate Nash equilibrium that induces a partition of the player set into relatively few societies, each consisting of similar individuals playing similar strategies. In this paper we characterize a family of games admitting the existence of such equilibrium. We also introduce the concept of ‘crowding types’ into our description of players and distinguish between the crowding type of a player—those characteristics of a player that have direct effects on others—and his tastes. With assumptions of ‘within crowding type anonymity’ and ‘linearity of taste-types’ we show that the number of societies can be uniformly bounded.

J. Ely and D. Fudenberg and D. K. Levine, 2005

In traditional reputation models, the ability to build a reputation is good for the long-run player. In [Ely, J., Valimaki, J., 2003. Bad reputation. NAJ Econ. 4, 2; http://www.najecon.org/v4.htm. Quart. J. Econ. 118 (2003) 785–814], Ely and Valimaki give an example in which reputation is unambiguously bad. This paper characterizes a class of games in which that insight holds. The key to bad reputation is that participation is optional for the short-run players, and that every action of the long-run player that makes the short-run players want to participate has a chance of being interpreted as a signal that the long-run player is “bad.”We allow a broad set of commitment types, allowing many types, including the “Stackelberg type” used to prove positive results on reputation. Although reputation need not be bad if the probability of the Stackelberg type is too high, the relative probability of the Stackelberg type can be high when all commitment types are unlikely.

D. Krahmer, 2006

The paper studies a repeated contest when contestants are uncertain about their true relative abilities. When ability and effort are complements, a favorable belief about one’s own ability stimulates effort and increases the likelihood of success. Success, in turn, reinforces favorable beliefs. We show that this implies that with positive probability players fail to learn their true relative abilities in equilibrium, and one player wins the contest with high probability forever. In this case, the prevailing player may be the actually worse player, and persistent inequality arises. We discuss some features of the model when the complementarity assumption is dropped.

J. Arifovic and R. D. McKelvey and S. Pevnitskaya, 2006

We report on a design of a Turing tournament and its initial implementation to learning in repeated 2- person games. The principal objectives of the tournament, named after the original Turing Test, are (1) to find learning algorithms (emulators) that most closely simulate human behavior, (2) to find algorithms (detectors) that most accurately distinguish between humans and machines, and (3) to provide a demonstration of how to implement this methodology for evaluating models of human behavior. In order to test our concept, we developed the software and implemented a number of learning models well known in the literature and developed a few detectors. This initial implementation found significant differences in data generated by these learning models and humans, with the greatest ones in coordination games. Finally, we investigate the stability of our result with respect to different evaluation approaches.

I. Milchtaich, 2006

Different kinds of networks, such as transportation, communication, computer, and supply networks, are susceptible to similar kinds of inefficiencies. These arise when congestion externalities make the cost for each user depend on the other users’ choice of routes. If each user chooses the least expensive (e.g., the fastest) route from the users’ common point of origin to the common destination, the result may be Pareto inefficient in that an alternative choice of routes would reduce the costs for all users. Braess’s paradox represents an extreme kind of inefficiency, in which the equilibrium costs may be reduced by raising the cost curves. As this paper shows, this paradox occurs in an (undirected) two-terminal network if and only if it is not series-parallel. More generally, Pareto inefficient equilibria occur in a network if and only if one of three simple networks is embedded in it.

C.D. Aliprantis and G. Camera and D. Puzzellob, 2006

We develop theoretical underpinnings of pairwise random matching processes. We formalize the mechanics of matching, and study the links between properties of the different processes and trade frictions. A particular emphasis is placed on providing a mapping between matching technologies and informational constraints.

A. Cassar, 2006

A laboratory experiment has been designed to study coordination and cooperation in games played on local, random and small-world networks. For the coordination game, the results revealed a tendency for coordination on the payoff-dominant equilibrium in all three networks, but the frequency of payoff-dominant choices was significantly higher in small-world networks than in local and random networks. For the prisoner’s dilemma game, cooperation was hard to reach on all three networks, with average cooperation lower in small-world networks than in random and local networks. Two graph-theoretic characteristics—clustering coefficient and characteristic path length—exhibited a significant effect on individual behavior, possibly explaining why the small-world network, with its high clustering coefficient and short path length, is the architecture of relations that drive a system towards equilibrium at the quickest pace.

D. K. Levine and W. Pesendorfer

We study evolutionarily stable outcomes for a class of games that admit cooperation and conflict as possible Nash equilibria. We make use of two ideas: existing strategies are more likely to be imitated than new strategies are to be introduced; players are able to identify opponents’ behavior prior to interaction. The long-run evolutionary limit is efficient for the case of perfect recognition of opponents’ behavior. For the case of imperfect recognition, efficiency is not achieved and long-run outcomes are more efficient the more accurate is the information. Strategies that emerge in the long run are those where players reward opponents who are likely to play the same way, and punish opponents who are likely to play differently.

S. M. Manson

This paper examines the use of evolutionary programming in agent-based modelling to implement the theory of bounded rationality. Evolutionary programming, which draws on Darwinian analogues of computing to create software programs, is a readily accepted means for solving complex computational problems. Evolutionary programming is also increasingly used to develop problem-solving strategies in accordance with bounded rationality, which addresses features of human decision-making such as cognitive limits, learning, and innovation. There remain many unanswered methodological and conceptual questions about the linkages between bounded rationality and evolutionary programming. This paper reports on how changing parameters in one variant of evolutionary programming, genetic programming, affects the representation of bounded rationality in software agents. Of particular interest are: the ability of agents to solve problems; limits to the complexity of agent strategies; the computational resources with which agents create, maintain, or expand strategies; and the extent to which agents balance exploration of new strategies and exploitation of old strategies.

Games and Economic Behaviour

IJGIS special issue on spatial agent-based modelling 2006

Journal of Artificial Societies and Social Simulation

Riccardo Boero - The Spatial Dimension and Social Simulations: A Review of Three Books. Um texto interessante e uma boa revisao de tres livros tratando de sistemas de agentes e da importancia da dimensao espacial. The Spatial Dimension and Social Simulations

Caípitulo 2 do Russel sobre agentes.

CASA - Centre for Advanced Spatial Analysis (http://www.casa.ucl.ac.uk/news/index.htm).

Vale uma olhadela no site deste cara. E professor da USP Poli com interfaces com Portugal e Franca e no Brasil na area de Multi-agentes.

http://www.pcs.usp.br/~jaime/#projetos

with Suresh Naidu: Institutional Equilibrium Selection by Intentional Idiosyncratic Play, 2004

with Hebert Gintis: The inheritance of inequality, 2002