A programming language for probabilistic computation Seminars/Workshops

Abstract

As probabilistic computations play an increasing role in solving various problems, researchers have designed probabilistic languages which treat probability distributions as primitive datatypes. Most probabilistic languages, however, focus only on discrete distributions and have limited expressive power. In this talk, I will present a probabilistic language, called PTP, which uniformly supports all kinds of probability distributions -- discrete distributions, continuous distributions, and even those belonging to neither group. Its mathematical basis is sampling functions, i.e., mappings from the unit interval (0.0,1.0] to probability domains. The practicality of PTP is demonstrated with three applications in robotics: robot localization, people tracking, and robotic mapping. The development of PTP is an effort to marry, in one of many possible ways, two seemingly unrelated disciplines: programming language theory and robotics. I will briefly talk about other problems in programming language theory that are relevant to robotics, such as distributed computation, dimension analysis, and parameter learning.