An Abductive Framework for Horn Knowledge Base Dynamics
This work addresses the challenge of updating belief states in AI systems, particularly for domains requiring flexible and immutable knowledge representation, but it appears incremental as it builds on existing dyadic representation approaches.
The paper tackles the problem of belief and knowledge dynamics in autonomous systems by introducing an abductive framework for Horn knowledge base dynamics, which handles non-deductively closed belief states and immutable parts by treating the immutable part as a new logical system with its own consequence relation and closure operator.
The dynamics of belief and knowledge is one of the major components of any autonomous system that should be able to incorporate new pieces of information. We introduced the Horn knowledge base dynamics to deal with two important points: first, to handle belief states that need not be deductively closed; and the second point is the ability to declare certain parts of the belief as immutable. In this paper, we address another, radically new approach to this problem. This approach is very close to the Hansson's dyadic representation of belief. Here, we consider the immutable part as defining a new logical system. By a logical system, we mean that it defines its own consequence relation and closure operator. Based on this, we provide an abductive framework for Horn knowledge base dynamics.