Computing Probabilistic Bisimilarity Distances for Probabilistic Automata

@article{Bacci2021ComputingPB,
  title={Computing Probabilistic Bisimilarity Distances for Probabilistic Automata},
  author={Giorgio Bacci and Giovanni Bacci and Kim G. Larsen and Radu Mardare and Qiyi Tang and Franck van Breugel},
  journal={Logical Methods in Computer Science},
  year={2021},
  volume={17},
  pages={1-36}
}
The probabilistic bisimilarity distance of Deng et al. has been proposed as a robust quantitative generalization of Segala and Lynch's probabilistic bisimilarity for probabilistic automata. In this paper, we present a novel characterization of the bisimilarity distance as the solution of a simple stochastic game. The characterization gives us an algorithm to compute the distances by applying Condon's simple policy iteration on these games. The correctness of Condon's approach, however, relies… Expand
1 Citations

Figures and Tables from this paper

Computing Probabilistic Bisimilarity Distances for Probabilistic Automata
TLDR
The characterization of the bisimilarity distance is presented as the solution of a simple stochastic game and an algorithm to compute the distances by applying Condon's simple policy iteration on these games is presented. Expand

References

SHOWING 1-10 OF 51 REFERENCES
Computing Probabilistic Bisimilarity Distances for Probabilistic Automata
TLDR
The characterization of the bisimilarity distance is presented as the solution of a simple stochastic game and an algorithm to compute the distances by applying Condon's simple policy iteration on these games is presented. Expand
Deciding Probabilistic Bisimilarity Distance One for Probabilistic Automata
TLDR
This paper presents a polynomial time algorithm that decides distance one and gives an alternative characterization of the probabilistic bisimilarity distances as a basis for a policy iteration algorithm. Expand
On the Complexity of Computing Probabilistic Bisimilarity
TLDR
It is shown that the problem of computing probabilistic bisimilarity is P-hard by reduction from the monotone circuit value problem and that the discounted pseudometric is rational and can be computed exactly in polynomial time using the network simplex algorithm and the continued fraction algorithm. Expand
Approximating a Behavioural Pseudometric without Discount for Probabilistic Systems
TLDR
A key ingredient of the algorithm is Tarski's decision procedure for the first order theory over real closed fields, and by exploiting the Kantorovich-Rubinstein duality theorem it can restrict to the existential fragment for which more efficient decision procedures exist. Expand
Computing Probabilistic Bisimilarity Distances via Policy Iteration
TLDR
A modification of a basic version of an algorithm to compute the probabilistic bisimilarity distances for a labelled Markov chain is shown to be the policy iteration algorithm applied to the corresponding simple stochastic game. Expand
Approximate Analysis of Probabilistic Processes: Logic, Simulation and Games
TLDR
A new semantics to a known simple logic for probabilistic processes is defined and it is shown that it characterises a notion of epsi-simulation, and most of these notions can be extended to deal with Probabilistic systems that allow non-determinism as well. Expand
Algorithms for Game Metrics
TLDR
It is shown that checking whether the metric distance is bounded by a rational r, can be done via a reduction to the theory of real closed fields, involving a formula with three quantifier alternations, improving the previously known reduction. Expand
On-the-Fly Exact Computation of Bisimilarity Distances
TLDR
This paper proposes an efficient on-the-fly algorithm which computes exactly the distances between given states and avoids the exhaustive state space exploration, and improves the efficiency of the corresponding iterative algorithms with orders of magnitude. Expand
The Complexity of Computing a Bisimilarity Pseudometric on Probabilistic Automata
TLDR
A generalization of the classical game-theoretic characterization of bisimilarity, given in terms of simple stochastic games, plays a central role in the proofs. Expand
Algorithms for Game Metrics (Full Version)
TLDR
It is shown that checking whether the metric distance is bounded by a rational r, can be accomplished via a reduction to the theory of real closed fields, involving a formula with three quantifier alternations, yielding O(|G| 5)) time complexity, improving the previously known reduction. Expand
...
1
2
3
4
5
...