Catálogo de publicaciones - libros

The main motivation guiding our research during the last 15 years, formulated as question, has been: Is it possible to aid decision-makers or their advisors who want to prevent the outbreak of hostilities/wars or to end them by means of negotiations or mediation, by giving them (interactively) recommendations as the result of applying Artificial Intelligence methods to existing war/crisis/mediation databases?

Ultimatum bargaining models of international interactions suggest that when conflict is costly and the actors are fully informed, the probability of conflict goes to zero. However, conflict occurs with some positive probability when the challenger is uncertain about the defender’s reservation value. I employ a simple ultimatum game of bargaining to evaluate two traditional power-centric theories of world politics, balance of power and power transition theory. The formal and empirical analyses demonstrate that as states approach power parity, information asymmetries are greatest, thus enhancing the probability of militarized conflict. Uncertainty is a central cause of conflict emergence and is correlated with the distribution of observable capabilities. Recognizing the relationship between the distribution of power and uncertainty offers a more sophisticated interpretation of power-centric explanations of world politics.

The ability to see the world from another person’s perspective is critical for mutual understanding and conflict prevention and resolution. It allows us to understand others’ subjective perspectives, and to predict, to some extent, their reactions to particular situations or events. This then allows us to identify specific situations or actions which are likely to trigger desirable (or undesirable) behavior, and thereby aid in the prevention and resolution of potential conflicts. A key component of these capabilities is the understanding of the other individuals’ values and motivations, patterns of reasoning, and behavior determined by their personalities, and individual reactions that may occur as a result of specific emotions. This type of understanding in turn necessitates understanding how different emotions and traits influence decision-making, and how these influences may result in specific decision-making styles or biases. This chapter describes a generic methodology for representing the effects of multiple, interacting emotional states and personality traits on decision- making, and an associated computational cognitive architecture which implements this methodology. I present results of an evaluation experiment that demonstrates the architecture’s ability to model individual tactical decision- making and produce observable behavior differences resulting from distinct individual profiles. I then discuss how the methodology and architecture would be extended to model strategic, political decision-making, and how it could support a variety of activities geared towards international conflict prevention and resolution. I conclude with specific theoretical and pragmatic challenges associated with this approach to computer-aided conflict prevention and resolution.

This chapter sketches out some new ways to look at conflict data sets. Since political scientists have more computer power available to them than at perhaps any point in the past, the paper emphasizes methods whose principle feature is that they purchase substantive realism at the cost of more compute cycles, not more advanced statistics. Existing statistical theory is sufficient to perform much more realistic analyses than are typically performed, but it is not necessarily found in the standard location. Most of the models and methods described here can be found in other guises in the field of machine learning. Political methodology is often accused of importing techniques wholesale from other disciplines, particularly econometrics, and by introducing machine learning as another field worth mining, this paper continues a long tradition.

Our knowledge of physics, chemistry, engineering, even psychology have all been used to develop weapons and approaches for winning armed conflicts. Especially the information sciences have been developed with a tremendous amount of funding and much of the initial motivation stemming from military needs and later developed with its funding. Robert Trappl has posed to the wider research community a fascinating question: Have our technologies and scientific approaches grown to the point that they can be applied to the much more difficult question of supporting the peaceful resolution of conflicts among nations? He has dubbed such technology “peacefare”. Sometimes it is easier to design a complex system by starting with a solution and working backwards. Hence I wrote a story. Since I am a computer scientist and technologist, my story is about the technology that could provide some enhanced basis for analysis of growing international tensions and possible mediation. This story also is a way of eliciting from our broad research community our goals, our assumptions, and the success criteria for a system like “Peacemaker 2020.”