A Bayesian Approach to Model Checking Biological Systems
read more
Citations
Statistical model checking: an overview
Uppaal SMC tutorial
A Survey of Statistical Model Checking
Statistical model checking for cyber-physical systems
Bayesian statistical model checking with application to Simulink/Stateflow verification
References
Exact Stochastic Simulation of Coupled Chemical Reactions
Model checking
Theory of probability
The temporal logic of programs
Related Papers (5)
Frequently Asked Questions (13)
Q2. How can the authors incorporate prior knowledge into their algorithm?
the authors note that because the authors adopt a Bayesian approach, their algorithm can incorporate prior knowledge, in the form of a probability distribution, P (θ), when available.
Q3. What is the correct choice for summarizing the prior probability distribution in Statistical Model Checking?
the Beta distribution is the appropriate choice for summarizing the prior probability distribution in Statistical Model Checking.
Q4. What is the Bayesian Model Checking algorithm?
In their experiments, the Bayesian Model Checking algorithm used uniform priors, and accepted a hypothesis when it was 10000 times more likely than the other hypothesis (Bayes Factor threshold T = 10000).
Q5. How many samples did prism need to estimate the probability of the BLTL formulae being?
The statistical estimation engine of the PRISM model checker always needed 92042 samples to estimate the probability of the BLTL formulae being true.
Q6. How did the authors study the Bayesian Model Checking algorithm?
The authors also studied SBML models using the implementation of Gillespie’s Stochastic Simulation Algorithm in Matlab’s Systems Biology Toolbox.
Q7. What is the way to test a simple hypothesis?
It can be shown that the SPRT is optimal for simple hypothesis testing, in the sense that it minimizes the expected number of samples among all the tests satisfying the same Type The authorand II errors [43], when either H ′0 or H ′ 1 is true.
Q8. How can the Bayes factor be computed?
the authors see that the Bayes factor can be computed by means of standard, well-known numerical methods, thereby simplifying the implementation of the algorithm.
Q9. What is the first application of Bayesian Sequential Hypothesis Testing to statistical Model Checking?
The contributions of this paper are as follows: • The first application of Bayesian Sequential Hypothesis Testing to statisticalModel Checking,• The first hypothesis-testing based statistical Model Checking algorithm designed for composite hypotheses, which can in particular include prior knowledge via a mixture of prior distributions, • A theorem proving that their algorithm terminates with probability 1, • Error bounds for their algorithm, and • A series of case studies using Systems Biology models demonstrating that ourmethod is empirically more efficient than existing algorithms for statistical Model Checking.
Q10. What is the performance of the Bayesian Model Checking algorithm?
(ii) The performance of both the Wald’s algorithm [42] and their Bayesian Model Checking algorithm degrades as the threshold probability (i.e., θ) in the PBLTL temporal logic formula gets close to the actual probability of the model satisfying the BLTL formula.
Q11. What is the advantage of Bayesian priors in Systems Biology?
This advantage in efficiency is important in the context of Systems Biology as the cost of generating traces is not necessarily negligible.
Q12. What is the syntax of the logic given by Wald’s SPRT?
The syntax of the logic is given by the following grammar:φ ::= x∼v | (φ1 ∨ φ2) | (φ1 ∧ φ2) | ¬φ1 | (φ1Utφ2), where ∼ ∈ {≥,≤,=}, x ∈ SV , v ∈ Q, and t ∈ Q≥0.
Q13. What is the probability of the number of bound Cyclin molecules?
The authors check the property that the probability of the number of bound Cyclin molecules exceeds 3 units within 0.5 time units exceeds θ (for various values of θ):H0 : M |= P≥θ[ F0.5 (cyclin bound > 3) ]