Abstract not found

and complexity

“A compromise is the art of dividing a cake in such a way that everyone believes he has the biggest piece.” Ludwig Erhard (1897–1977) The problem of fairly dividing a cake (as a metaphor for a heterogeneous divisible good) has been the subject of much interest since the 1940’s, and is of importance in multiagent resource allocation. Two fairness criteria are usually considered: proportionality, in the sense that each of the n agents receives at least 1/n of the cake; and the stronger property of envy-freeness, namely that each agent prefers its own piece of cake to the others’ pieces. For proportional division, there are algorithms that require O(n log n) steps, and recent lower bounds imply that one cannot do better. In stark contrast, known (discrete) algorithms for envy-free division require an unbounded number of steps, even when there are only four agents. In this paper, we give an Ω(n2) lower bound for the number of steps required by envy-free cake-cutting algorithms. This result provides, for the first time, a true separation between envy-free and proportional division, thus giving a partial explanation for the notorious difficulty of the former problem. 1

∗ A preliminary version of this paper has been presented at the 1st DIMACS-LAMSADE

Sensor networks arise as one of the most promising technologies for the next decades. The recent emergence of small and inexpensive sensors based upon microelectromechanical system (MEMS) ease the development and proliferation of this kind of networks in a wide range of real-world applications. Multi-Agent systems (MAS) have been identified as one of the most suitable technologies to contribute to this domain due to their appropriateness for modeling autonomous self-aware sensors in a flexible way. Firstly, this survey summarizes the actual challenges and research areas concerning sensor networks while identifying the most relevant MAS contributions. Secondly, we propose a taxonomy for sensor networks that classifies them depending on their features (and the research problems they pose). Finally, we identify some open future research directions and opportunities for MAS research. 1.

In a survey on the theory and practice of agent system deployment, conducted by the AgentLink workgroup on networked agents, it was found that there are an increasing number of initiatives for the migration of agents research towards new Internet technologies such as the semantic web, Grid, and Web services. In fact, Grid computing and multi-agent systems research have similar objectives. They both aim to achieve “large-scale open distributed systems, capable of being able to effectively and dynamically deploy and redeploy computational (and other) resources as required, to solve computationally complex problems ” [Foster and Kesselman 2003]. On the one hand, service-oriented Grid architectures need to support dynamic cooperation, negotiation, and adaptive interactions between Web services controlling Grid resources for efficient resource and task allocation and execution. On the other hand, the Grid can facilitate agent communication, life-cycle management, and access to resources for agents. Although the relevance of Grid for agent research and vice versa has been identified in several forums, actual collaborative applications are still in their infancy. In this article, we discuss our recent work on deploying multi-agent negotiation techniques to facilitate dynamic negotiation for Grid resources as a step closer to an adaptive and autonomous Grid. In particular, we describe a Web service development of the Contract Net Protocol

This paper describes the results of two joint academy-industry

Abstract not found

We consider a multiagent resource allocation domain where the marginal production of each resource is diminishing. A set of identical, self-interested agents requires access to sharable resources in the domain. We present a distributed and random allocation procedure, and demonstrate that the allocation converges to the optimal in terms of utilitarian social welfare. The procedure is based on direct interaction among the agents and resource owners (without the use of a central authority). We then consider potential strategic behavior of the selfinterested agents and resource owners, and show that when both act rationally and the domain is highly competitive for the resource owners, the convergence result still holds. The optimal allocation is arrived at quickly; given a setting with k resources and n agents, we demonstrate that the expected number of timesteps to convergence is O(k ln n), even in the worst case, where the optimal allocation is extremely unbalanced. Our allocation procedure has advantages over a mechanism design approach based on Vickrey-Clarke-Groves (VCG) mechanisms: it does not require the existence of a central trusted authority, and it fully distributes the utility obtained by the agents and resource owners (i.e., it is strongly budgetbalanced).

Cake cutting is a playful name for the problem of fairly dividing a heterogeneous divisible good among a set of agents. The agent valuations for different pieces of cake are typically assumed to be additive. However, in certain practical settings this assumption is invalid because agents may not have positive value for arbitrarily small “crumbs” of cake. In this paper, we propose a new, more expressive model of agent valuations that captures this feature. We present an approximately proportional algorithm for any number of agents that have such expressive valuations. The algorithm is optimal in the sense that no other algorithm can guarantee a greater worst-case degree of proportionality. We also design an optimal approximately proportional and fully envy-free algorithm for two agents. 1