Compositional Stochastic Modeling and Probabilistic Programming

2012/12/03 by Eric Mjolsness, Mjolsness, Eric
Computer Science · #Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Programming Languages (cs.PL) #cs.AI #cs.PL

paper · pdf · doi:10.48550/arxiv.1212.0582

Extended Abstract for the Neural Information Processing Systems (NIPS) Workshop on Probabilistic Programming, 2012

arxiv created 2012/12/03 · arxiv updated 2012/12/05

Abstract

Probabilistic programming is related to a compositional approach to stochastic modeling by switching from discrete to continuous time dynamics. In continuous time, an operator-algebra semantics is available in which processes proceeding in parallel (and possibly interacting) have summed time-evolution operators. From this foundation, algorithms for simulation, inference and model reduction may be systematically derived. The useful consequences are potentially far-reaching in computational science, machine learning and beyond. Hybrid compositional stochastic modeling/probabilistic programming approaches may also be possible.

Related

• An Introduction to Probabilistic Programming
• Compositional Abstraction-based Synthesis of General MDPs via Approximate Probabilistic Relations
• Stochastic Probabilistic Programs
• A New Approach to Probabilistic Programming Inference
• Church: a language for generative models