This paper presents an algorithm for extract ing propositions from trained neural networks. The algorithm is a decompositional approach which can be applied to any neural networ...
Although it is acknowledged that multi-way dataflow constraints are useful in interactive applications, concerns about their tractability have hindered their acceptance. Certain l...
Max and min operations have interesting properties that facilitate the exchange of information between the symbolic and real-valued domains. As such, neural networks that employ m...
The AI literature contains many definitions of diagnostic reasoning most of which are defined in terms of the logical entailment relation. We use existing work on approximate en...
It is generally accepted that knowledge based systems would be smarter and more robust if they can manage inconsistent, incomplete or imprecise knowledge. This paper is about a fo...
The paper proposes a set of principles and a general architecture that may explain how language and meaning may originate and complexify in a group of physically grounded distribu...
Motivated by the need to reason about utilities, and inspired by the success of bayesian networks in representing and reasoning about probabilities, we introduce the notion of uti...
This paper presents an interactive method for building a controller for dynamic systems by using a combination of knowledge acquisition and machine learning techniques. The aim is...
cal maps provide a useful abstraction for robotic navigation and planning. Although stochastic mapscan theoreticallybe learned using the Baum-Welch algorithm,without strong prior ...
The past several years have seen much progress in the area of propositional reasoning and satisfiability testing. There is a growing consensus by researchers on the key technical ...