Fine-Tuning the Odds in Bayesian Networks

  title={Fine-Tuning the Odds in Bayesian Networks},
  author={Bahar Salmani and Joost-Pieter Katoen},
. This paper proposes new analysis techniques for Bayes networks in which conditional probability tables (CPTs) may contain symbolic variables. The key idea is to exploit scalable and powerful techniques for synthesis problems in parametric Markov chains. Our techniques are applicable to arbitrarily many, possibly dependent, parameters that may occur in multiple CPTs. This lifts the severe restrictions on parameters, e.g., by restricting the number of parametrized CPTs to one or two, or by… 

Parameter Synthesis in Markov Models: A Gentle Survey

The main ideas underlying state-of-the-art algorithms that established an impressive leap over the last decade enabling the fully automated analysis of models with millions of states and thousands of parameters are described.

Abstraction-Refinement for Hierarchical Probabilistic Models

The key ideas to accelerate analysis of Markov decision processes are to treat the behavior of the subroutine as uncertain and only remove this uncertainty by a detailed analysis if needed, and to abstract similar subroutines into a parametric template, and then analyse this template.

Model Checking Finite-Horizon Markov Chains with Probabilistic Inference

The symbolic verification of Markov chains with respect to finite horizon reachability properties is revisit and Rubicon, a tool that transpiles Prism models to the probabilistic inference tool Dice, is developed as a first step towards integrating both perspectives.



Parameter Synthesis for Markov Models: Faster Than Ever

The technique provides the first sound and feasible method for performing parameter synthesis of Markov decision processes and outperforms the existing analysis of parametric Markov chains by several orders of magnitude regarding both run-time and scalability.

PROPhESY: A PRObabilistic ParamEter SYnthesis Tool

ProPhESY, a tool for analyzing parametric Markov chains (MCs), can compute a rational function (i.e., a fraction of two polynomials in the model parameters) for reachability and expected reward objectives and supports the novel feature of conditional probabilities.

Computing Conditional Probabilities in Markovian Models Efficiently

For Markov decision processes, it is shown that the task to compute maximal reachability probabilities under reachability conditions is solvable in polynomial time, while it was conjectured to be computationally hard.

Bayesian Inference by Symbolic Model Checking

A simple translation from Bayesian networks into tree-like Markov chains such that inference can be reduced to computing reachability probabilities using probabilistic sentential decision diagrams and vtrees, a scalable symbolic technique in AI inference tools.

Finding Provably Optimal Markov Chains

This paper proposes to tackle ETR-hard problems by a tight combination of two significantly different techniques: monotonicity checking and parameter lifting, which is an abstraction technique based on the iterative evaluation of pMCs without parameter dependencies.

Fast Parametric Model Checking through Model Fragmentation

Parametric model checking (PMC) computes algebraic formulae that express key non-functional properties of a system (reliability, performance, etc.) as rational functions of the system and environment

Analysis of Bayesian Networks via Prob-Solvable Loops

It is shown that various BNs, such as discrete, Gaussian, conditional linear Gaussian and dynamic BNs), can be naturally encoded as Prob-solvable loops, and can automatically solve several BN related problems, including exact inference, sensitivity analysis, filtering and computing the expected number of rejecting samples in sampling-based procedures.

Are Parametric Markov Chains Monotonic?

A simple algorithm to check whether reachability probabilities in parametric Markov chains are monotonic in (some of) the parameters is presented, which cheaply checks a sufficient condition for monotonicity.

On the Complexity of Reachability in Parametric Markov Decision Processes

The complexity of finding values for parameters such that the induced MDP satisfies some reachability constraints is studied, and all known lower bounds are improved and ETR-completeness results for distinct variants of this problem are provided.