2019/05/30 by Natan T’Joens, T'Joens, Natan, Thomas Krak +5
Biochemistry, Genetics and Molecular Biology · Computer Science · #Advanced Database Systems and Queries #Bayesian Modeling and Causal Inference #Data Management and Algorithms #FOS: Mathematics #Formal Methods in Verification #Gene Regulatory Network Analysis #Probability (math.PR)
paper · pdf · doi:10.48550/arxiv.1905.12968
openalex publication_date 2019/05/30 · openalex created_date 2022/07/29 · openalex updated_date 2026/07/28
We present an algorithm that can efficiently compute a broad class of\ninferences for discrete-time imprecise Markov chains, a generalised type of\nMarkov chains that allows one to take into account partially specified\nprobabilities and other types of model uncertainty. The class of inferences\nthat we consider contains, as special cases, tight lower and upper bounds on\nexpected hitting times, on hitting probabilities and on expectations of\nfunctions that are a sum or product of simpler ones. Our algorithm exploits the\nspecific structure that is inherent in all these inferences: they admit a\ngeneral recursive decomposition. This allows us to achieve a computational\ncomplexity that scales linearly in the number of time points on which the\ninference depends, instead of the exponential scaling that is typical for a\nnaive approach.\n