There are two foci of this proposal. First, research on fundamental models of distributed computing will be continued. The foundations of communication protocols will be investigated using game automata, logics of knowledge and time, and other suitable tools. The long range goal is to help develop mathematical models and theories for protocols which will have the same footing as the current accepted mathematical models and theories have for sequential and parallel models. The second focus is to begin a new research effort into the connectionist models of computation. There is hope, and even some evidence, that the new connectionist models will revolutionize the way we compute. There are a number of obvious differences between traditional models of computation and connectionist models. The goal of this research is to investigate how deep these differences are. A comparison of the traditional parallel models of computation and the connectionist models will be used to determine if indeed the connectionist models are an improvement over the traditional ones.
