Piotr J. Gmytrasiewicz. On reasoning about other agents. In M. Wooldridge, J. P. Muller, and M. Tambe, editors, Intelligent Agents Volume II, Lecture Notes in Artificial Intelligence, pages 143--155. Springer-Verlag, 1996.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....the e#ects of increased agent modeling capabilities within our economic model, we defined a set of techniques that our agents can use for learning and using models. We divide the agents into classes that correspond to their modeling capabilities. The hierarchy we present is inspired by RMM [6], but is functionbased rather than matrix based, and includes learning. We start with agents with no models (also referred to as 0 level agents) who must base their actions purely on their inputs and the rewards they receive. They are not aware that there are other agents out there. Agents with ....

Piotr J. Gmytrasiewicz. On reasoning about other agents. In M. Wooldridge, J. P. Muller, and M. Tambe, editors, Intelligent Agents Volume II, Lecture Notes in Artificial Intelligence, pages 143--155. Springer-Verlag, 1996.

....other agents[9] A second area of related work is research specifically focused on agent modeling and plan recognition. Section 1 has discussed some of this work. In addition, some formal approaches for agent modeling, and in particular for recursive agent modeling, are also being investigated[11]. Vidal and Durfee attack the problem of combinatorial explosion in such recursive modeling, and propose a formal approach to tame the combinatorics[31] Understanding the relationship of these formal approaches to approaches inspired by practical applications, as in the work presented in this ....

P. Gmytrasiewicz. On reasoning about other agents. In M. Wooldridge, J. Muller, and M. Tambe, editors, Intelligent Agents, Vol II -- Proceedings of the 1995 workshop on Agent theories, Architecturesand Languages(ATAL-95), Lectures Notes in Articificial Intelligence. Springer-Verlag, Heidelberg, 1996. (In this volume).

....the effects of increased agent modeling capabilities within our economic model, we defined a set of techniques that our agents can use for learning and using models. We divide the agents into classes that correspond to their modeling capabilities. The hierarchy we present is inspired by RMM [7], but is functionbased rather than matrix based, and includes learning. We start with agents with no models (also referred to as 0 level agents) who must base their actions purely on their inputs and the rewards they receive. They are not aware that there are other agents out there. Agents with ....

Piotr J. Gmytrasiewicz. On reasoning about other agents. In M. Wooldridge, J. P. Muller, and M. Tambe, editors, Intelligent Agents Volume II --- Proceedings of the 1995.

....presented by Sen and Arora [9] who propose a scheme for learning opponent action probabilities and a maximum expected utility strategy for exploiting weaker opponents. Tambe et al. [11] have proposed an approach for tracking recursive agent models based on a plan recognition task. Gmytrasiewicz [4] has presented the Recursive Modeling Method (RMM) which uses nested models of other agents, combining game theoretic and decision theoretic mechanisms. Suryadi and Gmytrasiewicz [10] have proposed the use of influence diagrams for learning models about other agents. Vidal and Durfee [12] have ....

P.J. Gmytrasiewicz. On reasoning about other agents. In Intelligent Agents II, Lecture Notes in Artificial Intelligence (LNAI 1037). Springer Verlag, 1996.

....or non communicating situations. In competitive settings it is important to model others strategies in order to perform better against them [2] Strategies can be represented in terms of game theory [16] A decision theoretic approach has been also taken in the Recursive Modeling Method [10]. Finite automata has been used to represent rationalbounded strategies [12] Some proposals model the others plans [20] Concerning the model construction method, the simplest case arrives in cooperative settings, where honest agents tell the others what their characteristics are [4] ....

....the other agents nor how agents reason in order to achieve a solution. This approach takes an external perspective and analyzes the problem of all agents together in order to provide overall efficiency and stability by means of negotiation protocols (see [17] for further details) Gmytrasiewicz [10] has presented a research work that combines game theory with decision theoretic mechanisms. Assuming that agents are rational and the common knowledge of their payoff functions, he presents a decision theoretic mechanism that aims to let agents be able to reason about nested models about the ....

P.J. Gmytrasiewicz. On reasoning about other agents. In Intelligent Agents II, Lecture Notes in Artificial Intelligence (LNAI 1037). Springer Verlag, 1996.

....to intercept multiple incoming threats (as in anti air defense) but we believe that lessons learned in this domain generalize to other multi agent domains. 2 Background and Related Work Our prior work on deliberative decision theoretic method includes the Recursive Modeling Method (RMM) Gmytrasiewicz, 1996; Gmytrasiewicz et al. 1998; Noh and Gmytrasiewicz, 1997; 1998 ] We have implemented a full blown version of RMM which allows an agent to compute its best action given what is known about the other agents and about their states of knowledge and capabilities. In the task of coordinating agents ....

....with only long range interceptors, having only short range interceptors, and incapacitated, are depicted as the second level models in Figure 5, with their associated probabilities, in this example case 0:3, 0:3, 0:3, and 0:1, respectively. The Recursive Modeling Method uses dynamic programming [ Gmytrasiewicz, 1996; Noh and Gmytrasiewicz, 1997 ] to process model structures as in Figure 5 and determine the rational choice of coordinated action. In this case, Battery1 computes that if Battery2 is fully operational then the probability distribution over Battery2 s actions A, B, and S is [0:03; 0:97; 0:0] If ....

P. J. Gmytrasiewicz. On reasoning about other agents. In Intelligent Agents II: Agent Theories, Architectures, and Languages, pages 143-- 155, Berlin, 1996. Springer.

....the hostile missiles. Based on these attributes combined, each unit has to determine the optimal action from his probabilistic decision model. For the purpose of coordinated decision making in a multiagent environment, our research uses the Recursive Modeling Method (RMM) previously reported in [2, 3]. RMM enables an agent to model the other agents and to rationally coordinate with them even if no protocol or overall plan can be established explicitly in advance. Using RMM as a decision making tool, an agent rationally selects his action under uncertainty guided by the principle of expected ....

....by independent defense units. 3 Decision Theoretic Agent To be rational in decision theoretic sense, the agents follow the principle of maximum expected utility (PMEU) 10] In this section, we will show how PMEU can be implemented in this case study using the Recursive Modeling Method (RMM) RMM [2, 3] will be used to model the other agent, and to select the most appropriate missile to intercept by a given defense battery. 3.1 An Example Scenario Our approach is to take the agent oriented perspective. In the examples scenario (Fig. 1) we view the decision making through the eyes of an ....

[Article contains additional citation context not shown here]

P. J. Gmytrasiewicz. On reasoning about other agents. In Intelligent Agents II: Agent Theories, Architectures, and Languages, pages 143--155. Springer, 1996.

....and execute his action, and that coordination among the units is to emerge on the fly as the result of the units individual rational actions. For the purpose of coordinated decision making in a multiagent environment, our research uses the Recursive Modeling Method (RMM) previously reported in [1, 2]. RMM enables an agent to model the other agents and to rationally coordinate with them even if no protocol or overall plan can be established explicitly in advance. Using RMM as a decision making tool, an agent rationally selects his action under uncertainty guided by the principle of expected ....

....information about Battery2 s action. We call the last model the No info model. In RMM, each of the alternative models is assigned a probability indicating the likelihood of its correctness. Figure 1 is the Battery1 s model structure of depth two for the example scenario. Level 1: Level 2: No info [0,0,0,0,0,1] [1,0,0,0,0,0] No info [0,0,0,0,0,1] 1,0,0,0,0,0] Belief: A A B C C B D E F D E F Battery1 Battery2 A A B C C B D E F D E F Battery2 Battery1 A A B C C B D E F Battery1 1 1 28.0 42.9 41.3 50.8 52.4 55.9 42.6 23.6 38.5 47.9 49.5 53.0 39.6 37.1 19.4 44.9 46.5 50.0 44.8 42.2 40.6 ....

[Article contains additional citation context not shown here]

P. J. Gmytrasiewicz. On reasoning about other agents. In M. Wooldridge, J. P. Muller, and M. Tambe, editors, Intelligent Agents II: Agent Theories, Architectures, and Languages, pages 143--155, Berlin, 1996. Springer.

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