Sandholm T.W; Lesser V.R, Coalition formation among bounded rational agents, Proceedings of the Fourteenth International Joint Conference on AI, 1995, pag. 662-669  Home/Search   Document Details and Download   Summary   Related Articles   Check

....those obtained using cooperative reward, those obtained using sel sh reward and those which show emergent behaviour. Keywords: Multi Agent Systems, emergent behaviour, XCS, Learning Classi er Systems, El Farol Bar problem. 1 Introduction In this paper, MAXCS a Multi Agent system (MAS) 16][31] that learns using XCS [39] is used for social modelling on the El Farol Bar problem [3] A Multi agent System is a set of individual entities that can partially perceive their environment and interact autonomously with it. The agents learn from the rewards obtained from the environment. The ....

T. Sandholm and V.R. Lesser. Coalition formation among bounded rational agents. Technical Report 95-71, University of Massachusetts at Amherst, 1995.

....performance by acting in teams. When implementing a mechanism, one must fully understand the consequences of cooperative or collusive behavior. Team analysis can also give bounds on performance of agents in non cooperative games. Thus, coalition formation has been a rich area for investigation [8, 12]. Cooperative phenomena depend on the structure of the game being played. Our goal is to investigate the e#ects of coalition formation in the regime of the proportionally fair divisible auction to answer the following questions. How does one obtain the value of a coalition What is the reaction ....

T. W. Sandholm V. R. Lesser. Coalition formation among bounded rational agents. In 14th International Joint Conference on Artificial Intelligence, pages 662--669, Montreal, Canada, August 1996.

....in cooperative environments were suggested (e.g. 11] 2] In [1] the problem is addressed for self interested agents, but in superadditive environments. In [10] solutions were proposed for non superadditive environments, but the value of each coalition is known. Sandholm and Lesser [9] present a coalition formation model for bounded rational agents and a general classification of coalition games. As in [9] we also allow for varying coalitional values, but provide the agents with heuristics that could be computed in polynomial time. However, we assume that time is costly, ....

....but in superadditive environments. In [10] solutions were proposed for non superadditive environments, but the value of each coalition is known. Sandholm and Lesser [9] present a coalition formation model for bounded rational agents and a general classification of coalition games. As in [9] we also allow for varying coalitional values, but provide the agents with heuristics that could be computed in polynomial time. However, we assume that time is costly, and that agents take the coalition formation time into consideration when deciding on whether to join a coalition. In ....

[Article contains additional citation context not shown here]

Sandholm, T., and Lesser, V. Coalition formation among bounded rational agents, Artificial Intelligence Journal 94(12, (1997), 99-137,

....agent, or set of agents. Thus, coalition formation is a very important activity in agent societies. Unfortunately, most research in this area assumes societies that are closed and where agents have unlimited computational resources. Some attempts to ease up these restrictions have been made [27, 20] and a promising general framework has been suggested by Johansson [19] It should be noted that there are some similarities between society and coalition formation, e.g. similar reasoning is required by the agents when deciding whether to join a coalition or a society, i.e. weighing the pros ....

Sandholm, T.W. and Lesser, V.R.: Coalition Formation among Bounded Rational Agents. In: IJCAI'95, 1995.

....often relaxes the optimality criteria and bound one or more of these parameters. For instance we can bound the depth that a minimax algorithm is allowed to search, or how many levels of nested models 19 of each other agents are allowed to maintain. 3.7. 1 Cost of computation Sandholm and Lesser [SL95b] investigated a model in which agents have bounded rationality. They bound the rationality by introducing a cost for computational resources. In their example, agents bene t from participating in larger coalitions since with more resources, better (cheaper) plans can be made. But it takes time to ....

Tuomas Sandholm and Victor Lesser. Coalition formation among bounded rational agents. In 14th International Joint Conference on Articial Intelligence (IJCAI-95), pages 662-669, Montreal, Canada, 1995. ftp://ftp.cs.umass.edu/pub/lesser/sandholm-ijcai95-coalition.ps.

....a group of societies, a new meta society emerges which can be viewed as a single entity acting in a coherent manner when reacting to events that occur in its environment. At different levels in this hierarchy of societies, different social attitudes can be found. The study of coalition formation [9][13] provides examples of how selfish agents can eventually coordinate their actions and form an atomistic society that interacts in an also atomistic meta society. Resource allocation in a single and homogeneous society is one of the major interest areas of research in multi agent community [6] ....

Tuomas W. Sandholm and Victor R. Lesser. Coalition formation among bounded rational agents. In 14th International Joint Conference on Artificial Intelligence (IJCAI-95), pages 662--669, Montereal, Canada, 1995.

....[5] Tan [79] Schmidhuber [64] Schaerf et al. 63] Zeng and Sycara [87, 86] Stone and Veloso [73] Schneider et al. 65] Wei [83] and Brauer and Wei [9] Others have pursued methods of coalition formation to optimize multiagent coordination. Examples are work done by Sandholm and Lesser [58], Sandholm et al. 60] Zlotkin and Rosenschein [89] and Shehory and Kraus [68] Tambe [77] proposes tracking behavior of other agents as an optimization technique. Rosin and Belew [55] and Matos and Sierra [47] adopt evolutionary methods for enhancing performance. Other optimization techniques ....

Sandholm, T., and Lesser, V., "Coalition Formation Among Bounded Rational Agents", Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, pp. 662-669, 1995.

....denotes a wider class of objects than solvable does. Apart from this somewhat philosophical remark, in this paper we shall always mean the two terms as synonymous. 1 of research, in fact, may be related on agents decision making models, where an incomplete list of works is [Syc89] and [SL95, SL97, SLA 99] plus the references cited in these papers. The scope of this article falls primarily into the domain of decision making models. More precisely, decision making models for coalition formation and groupwork. First, we de ne a communication protocol among a set of agents that ....

....[OR94, Ch. 12] However, it should made clear that to what formal extent is there any similarity is an interesting question that we leave open to future work. Proposition (5) provides a simple example of a process that Sandholm and Lesser called coalition structure generation (see e.g. SL95, p. 662] The proofs of Theorem (7) Theorem (8) and Theorem (16) are examples of limiting games, where there is a winning strategy for pairwise ( coordination on a set of agents with common knowledge. In particular, if two agents agree upon a convention on leadership, that is, they ....

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. In C. S. Mellish, editor, Proceedings of the Fourteenth International Joint Conference on Articial Intelligence (IJCAI-95), pages 662-669, San Mateo, CA, 1995. Morgan Kaufmann.

....bidding; and other self adapting behaviours. Since a BDI agent has its own cognitive model, modifications in its beliefs can cause change in behaviour. Plan generation, and hence decision making, are dependant on the agents own dynamic model of the environment. Automated coalition formation [19] allows both static and dynamic formation of BDI teams. Coalition formation could be done in many different ways to enable the achievement of mutual goals or the exchange of benefits. Flexible Capacity: Unrestricted unit populations tha t allow large increases and decreases in total unit ....

Sandholm, T., Lesser, V.: Coalition Formation Among Bounded Rational Agents. 14th International Joint Conference on Artificial Intelligence (IJCAI-95), Montreal, Canada, (1995) pp. 662-669.

....is also considered illegal. However, as with shills, it can be hard to detect buyer collusion, especially in online markets where bidders are virtual. In fact, Multi Agent Systems research has developed technologies that can efficiently form coalitions even among previously unknown parties [36] posing an additional threat to online retail auctions. As explained, online auctions are unnecessarily hostile to customers and offer no long term benefits to merchants. Essentially, they pit merchant against customer in price tug of wars. This is not the type of relationship merchants ....

T. Sandholm and V. Lesser. "Coalition Formation among Bounded Rational Agents." 14th International Joint Conference on Artificial Intelligence (IJCAI'95), Montreal, Canada, 1995.

....is a decreasing price auction that maximizes privacy, but is auctioneer friendly since it guarantees the purchase at the highest possible price. All schemes provoke various manipulations [GMM98,GM98b] like fake bidders (shills) sellers acting as auctioneers, or coalition formation by bidders [SL95a]. The Vickrey auction [Vic61,San95] is a non discriminative auction scheme where the winning bidder pays the price of the second highest bid. This policy creates disincentive for speculative bidding and maximizes consumer surplus because bidders are best off to submit bids reflecting their true ....

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. In Proc. of the 14th Int. Joint Conference on Artificial Intelligence, pages 662--669, Montreal, Canada, 1995.

....price. In [GM98a, GMM98, GM98b] it has been pointed out that auction protocols either favor the bidder or the seller, forcing buyers and sellers into price wars and provoking various manipulations like fake bidders (shills) sellers acting as auctioneers, or coalition formation by bidders [SL95] Also for the widely used Vickrey auction [Vic61] it has been shown that bidding the truth is not always the dominant strategy for an agent [San95] In order to overcome the shortcomings of bilateral negotiations and auctions, we propose to mediate conflicting interests with a neutral third ....

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. In Proc. of the 14th Int. Joint Conference on Artificial Intelligence, pages 662--669, Montreal, Canada, 1995.

No context found.

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. In Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, pages 662-- 669, Montreal, Canada, Aug. 1995. Extended version appeared as University of Massachusetts at Amherst, Computer Science Department technical report 95-71.

....make the first contract beneficial [23] Again, if no such contracts appear, the agent can decommit. In many practical automated contracting settings, agents are bounded rational for example because limited computation resources bound their capability to solve combinatorially complex problems [23, 20, 21, 19]. The very fact that an agent s computation is bounded induces uncertainty. For example, the value of a contract may only be probabilistically known to the 44 agent at contract time. The leveled commitment contracting protocol allows the agent to continue deliberation regarding the value of the ....

....events. A normative theory relating the performance profiles of the algorithms of bounded rational agents to the issues of this paper is also desirable. We have already taken initial steps towards relating the performance profiles and optimal negotiation actions in a coalition formation problem [20, 21]. Finally, the relationship between leveled commitment contracting and explicit contracts among more than two agents should be studied in more detail. ....

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. In Proc. 14th International Joint Conference on Artificial Intelligence (IJCAI-95), pages 662--669, Montreal, Canada, Aug. 1995.

....make the first contract beneficial [23] Again, if no such contracts appear, the agent can decommit. In many practical automated contracting settings, agents are bounded rational for example because limited computation resources bound their capability to solve combinatorially complex problems [23, 20, 21, 19]. The very fact that an agent s computation is bounded induces uncertainty. For example, the value of a contract may only be probabilistically known to the 44 agent at contract time. The leveled commitment contracting protocol allows the agent to continue deliberation regarding the value of the ....

....events. A normative theory relating the performance profiles of the algorithms of bounded rational agents to the issues of this paper is also desirable. We have already taken initial steps towards relating the performance profiles and optimal negotiation actions in a coalition formation problem [20, 21]. Finally, the relationship between leveled commitment contracting and explicit contracts among more than two agents should be studied in more detail. ....

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. Technical Report 95-71, University of Massachusetts at Amherst Computer Science Department, 1995. Extended version.

....hinders unfair renegotiation. 1 1 Introduction In cooperative distributed problem solving [ Durfee et al. 1989 ] the system designer imposes an interaction protocol and a strategy (a mapping from state history to actions; a way to use the protocol) for each agent. In multiagent systems [ Sandholm and Lesser, 1995c; 1995a; 1995b; Rosenschein and Zlotkin, 1994; Durfee et al. 1993; Kraus et al. 1992; Wellman, 1992 ] the agents are provided with an interaction protocol, but each agent may choose its own strategy. This allows the agents to be constructed by separate designers and or represent 1 Supported by ARPA ....

....negotiator should only agree to contracts that can be executed so that the opponent is not motivated to defect at any point of the exchange. Automated negotiation has been mostly studied with respect to ex ante rationality: what contracts seem desirable to the agents before they are carried out [ Sandholm and Lesser, 1995b; 1995c; 1995a; Sandholm, 1993; Rosenschein and Zlotkin, 1994; Kraus et al. 1992; Wellman, 1992; Durfee et al. 1993 ] We suggest that contracts should also fulfill the condition of ex post rationality: abiding to the contract should be desirable to the agents at each step of the carrying out of the ....

Tuomas W Sandholm and Victor R Lesser. Coalition formation among bounded rational agents. In Proc. 14th International Joint Conference on Artificial Intelligence (IJCAI95) , Montreal, Canada, August 1995.

No context found.

Sandholm T. and Lesser, V. 1995a. Coalition Formation among Bounded Rational Agents. In Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence (IJCAI-95), Montreal, Canada.

No context found.

Sandholm T.W; Lesser V.R, Coalition formation among bounded rational agents, Proceedings of the Fourteenth International Joint Conference on AI, 1995, pag. 662-669

No context found.

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. In Proc. of IJCAI-95, pages 662--669, Montr'eal, 1995.

No context found.

T. Sandholm and V. Lesser. "Coalition Formation among Bounded Rational Agents." 14th International Joint Conference on Artificial Intelligence (IJCAI'95), Montreal, Canada, 1995.

No context found.

T. W. Sandholm and V. R. Lesser. Coalition formation among bounded rational agents. In Proc. of IJCAI-95, pages 662--669, Montr'eal, 1995.

No context found.

Sandholm, T. and Lesser, V. "Coalition formation among bounded rational agents", Artificial Intelligence Journal, 94(12) , 1997, pp. 99-137.

No context found.

Sandholm, T. and Lesser, V.: Coalition Formation among Bounded Rational Agents. 14th International Joint Conference on Articial Intelligence (1995) 662-669

No context found.

Sandholm, T. and Lesser, V.: Coalition Formation among Bounded Rational Agents. 14th International Joint Conference on Articial Intelligence (1995) 662-669

No context found.

Tuomas W. Sandholm and Victor R. Lesser. Coalition formation among bounded rational agents. In Chris S. Mellish, editor, IJCAI-95,volume I, pages 662--669, Denver, Colorado, August 1995. International Joint Conferences on Artificial Intelligence, Inc., Professional Book Center.

First 50 documents  Next 50

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