Towards an Open Negotiation Architecture for Heterogeneous Agents

Abstract

Abstract. This paper presents the design of an open architecture for heterogeneous negotiating agents. Both the system level architecture as well as the architecture for negotiating agents are provided. The main contribution of this paper is that it derives a precisely specified interface from these architectures that facilitates an easy integration of heterogeneous agents into the overall negotiation framework. The interface is defined as a set of adapters that allows for various levels of integration of agents into the system architecture. The functionality provided by the system architecture depends on the number of adapters that are implemented and used to connect an agent to this architecture, ranging from functionality to conduct a bilateral negotiation to functionality for computing agent internal performance measures such as the quality of an opponent model. The architecture is used as the basis of a competitive testbed which allows us to study various negotiating agents. The design yields a flexible negotiation framework that facilitates negotiating different domains potentially using different protocols whereas no details of the internal negotiating agent structure are enforced. An application of the framework is illustrated by integrating two agents from the literature.

A Multi-Agent Environment for Negotiation

“…The redesigned version is implemented in the Java programming language that is supported my the majority of computer platforms. The current version of SAMIN implements the architecture proposed in [13]. Figure 10.1 illustrates this architecture.…”

“…The latter adapter is mainly available for analysis purposes and research. For more details about the adapters the reader is referred to [13].…”

“…Efficiency of a negotiation strategy can be significantly improved with information about the preferences of the opponent [33]. In the literature a number of learning techniques have been proposed to learn the opponent's preferences model from the offers exchanged in a single-shot negotiation, see e.g., [34,13]. In [33] it was show that a successful negotiation strategy should make use of an opponent model.…”

“…This knowledge can be used in the negotiation strategy to increase the efficiency of the agent's own offers and increase the chance of acceptance of its offer by the opponent. Models of the opponent's preferences and strategy are typically learned by the agent from the evidence, such as negotiation agreements achieved in the previous negotiations [33], and offers sent by the opponent in multiple sessions of singleshot negotiations [13,34,18]. The learning techniques used in the agent can depend on the types of the models chosen to represent the opponent's preferences and strategy.…”

“…Interfaces and adapters have been developed to make it easy to integrate agents developed by others into SAMIN, see [13]. A number of the state-of-the-art agents have found a place in SAMIN: ABMP [17], Bayesian agent [14], Bayesian Tit-forTat [12], FBM [29], Trade-off agent [6], QO agent [24], Random Walker [11].…”