Some practical issues on implementing distributed multi-agent reasoning systems

As intelligent systems are being applied to increasingly larger, open and more complex problem domains. These domains demand a systematic approach to handle the complexity in knowledge engineering. The needs include modularity in representation, distribution in computation, as well as coherence in inference. Multiply Sectioned Bayesian Networks provide a distributed multi-agent framework to address these needs. According to the framework, a large system is partitioned into subsystems and represented as a set of related Bayesian subnets. To ensure exact inference, partitioning of a large system into subsystems must follow a set of technical assumptions. In this paper, we propose a practical method that gives the answer of how to achieve concrescence of agentspsila believes in a distributed multi-agent system, instead of restricting them by an assumption of being consistent.