EXTENDING CAUSAL METAMODELING TO A NON-MARKOVIAN QUEUE

Pracheta Amaranath 1, Anant Bhide 1, David Jensen 1 and Peter J. Haas 1 1 Manning College of Information and Computer Sciences, University of Massachusetts Amherst, MA, USA · 2026

Read EXTENDING CAUSAL METAMODELING TO A NON-MARKOVIAN QUEUE by Pracheta Amaranath 1, Anant Bhide 1, David Jensen 1 and Peter J. Haas 1 1 Manning College of Information and Computer Sciences, University of Massachusetts Amherst, MA, USA (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-00795.txt.

Description

Amaranath and co-authors speed up answers from costly simulations in this 2026 paper. A metamodel mimics a simulation model so users can skip full runs. A prior class, modular dynamic Bayesian networks, can answer many probabilistic and causal queries from one trained model. But it worked only for Markovian systems, where the future rests on the present alone.

Queues with memory break that limit. The team extends the method to a non-Markovian queue. They mimic non-exponential wait-time laws with phase-type laws, which Markov tools can handle. New problems arise: how many phases to use, how to learn the parameters fast, and how to slice time. They give first fixes for each. On a G/M/1 queue, their metamodel answers queries with good accuracy, they report. Inference runs orders of magnitude faster than direct simulation. The work claims the first causal metamodeling method for non-Markovian systems.

Catalog record

CREATORPracheta Amaranath 1, Anant Bhide 1, David Jensen 1 and Peter J. Haas 1 1 Manning College of Information and Computer Sciences, University of Massachusetts Amherst, MA, USA
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE40 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.00795

Browse the full library · About the Tower of Babel