Who chooses the assumptions
David Poole · 2007
In this paper we show how a number of different formulations of nonmonotonic reasoning, probabilistic reasoning and design can be combined into a coherent logic-based abductive framework. This framework is based on allowing consistent assumptions to be used to prove a goal. Different frameworks are characterised by who chooses the assumptions, whether an adversary chooses the assumptions, nature chooses the assumptions, or one gets to choose whatever assumptions one likes. 1 Introduction In artificial intelligence over the last decade there has been much work in logic-based nonmonotonic, probabilistic and abductive reasoning (see e.g., papers in [17, 44, 26]). In this paper, we show how a particular simple form of abductive reasoning can gives a unifying theme to many seeming disparate reasoning schemes, for example Circumscription [24] and Bayesian This paper is to appear in P. O'Rorke (Ed.) Abductive Reasoning, MIT Press, 1994. y Scholar, Canadian Institute for Advanced Researc...