The long-run average payoff per transition (mean payoff) is the main tool for specifying the performance and dependability properties of discrete systems. The problem of constructing a controller (strategy) simultaneously optimizing several mean payoffs has been deeply studied for stochastic and game-theoretic models. One common issue of the constructed controllers is the instability of the mean payoffs, measured by the deviations of the average rewards per transition computed in a finite "window" sliding along a run. Unfortunately, the problem of simultaneously optimizing the mean payoffs under local stability constraints is computationally hard, and the existing works do not provide a practically usable algorithm even for non-stochastic models such as two-player games. In this paper, we design and evaluate the first efficient and scalable solution to this problem applicable to Markov decision processes.

Effective security patrol management is critical for ensuring safety in diverse environments such as art galleries, airports, and factories. The behavior of patrols in these situations can be modeled by patrolling games. They simulate the behavior of the patrol and adversary in the building, which is modeled as a graph of interconnected nodes representing rooms. The designers of algorithms solving the game face the problem of analyzing complex graph layouts with temporal dependencies. Therefore, appropriate visual support is crucial for them to work effectively. In this paper, we present a novel tool that helps the designers of patrolling games explore the outcomes of the proposed algorithms and approaches, evaluate their success rate, and propose modifications that can improve their solutions. Our tool offers an intuitive and interactive interface, featuring a detailed exploration of patrol routes and probabilities of taking them, simulation of patrols, and other requested features. In close collaboration with experts in designing patrolling games, we conducted three case studies demonstrating the usage and usefulness of our tool. The prototype of the tool, along with exemplary datasets, is available at https://gitlab.fi.muni.cz/formela/strategy-vizualizer.

Long-run average optimization problems for Markov decision processes (MDPs) require constructing policies with optimal steady-state behavior, i.e., optimal limit frequency of visits to the states. However, such policies may suffer from local instability in the sense that the frequency of states visited in a bounded time horizon along a run differs significantly from the limit frequency. In this work, we propose an efficient algorithmic solution to this problem.

We consider the problem of synthesizing resilient and stochastically stable strategies for systems of cooperating agents striving to minimize the expected time between consecutive visits to selected locations in a known environment. A strategy profile is resilient if it retains its functionality even if some of the agents fail, and stochastically stable if the visiting time variance is small. We design a novel specification language for objectives involving resilience and stochastic stability, and we show how to efficiently compute strategy profiles (for both autonomous and coordinated agents) optimizing these objectives. Our experiments show that our strategy synthesis algorithm can construct highly non-trivial and efficient strategy profiles for environments with general topology.

Consider oriented graph nodes requiring periodic visits by a service agent. The agent moves among the nodes and receives a payoff for each completed service task, depending on the time elapsed since the previous visit to a node. We consider the problem of finding a suitable schedule for the agent to maximize its long-run average payoff per time unit. We show that the problem of constructing an epsilon-optimal schedule is PSPACE-hard for every fixed non-negative epsilon, and that there exists an optimal periodic schedule of exponential length. We propose randomized finite-memory (RFM) schedules as a compact description of the agent's strategies and design an efficient algorithm for constructing RFM schedules. Furthermore, we construct deterministic periodic schedules by sampling from RFM schedules.

We consider the mobile robot path planning problem for a class of recurrent reachability objectives. These objectives are parameterized by the expected time needed to visit one position from another, the expected square of this time, and also the frequency of moves between two neighboring locations. We design an efficient strategy synthesis algorithm for recurrent reachability objectives and demonstrate its functionality on non-trivial instances.

We consider the problem of efficient patrolling strategy adaptation in a changing environment where the topology of Defender’s moves and the importance of guarded targets change unpredictably. The Defender must instantly switch to a new strategy optimized for the new environment, not disrupting the ongoing patrolling task, and the new strategy must be computed promptly under all circumstances. Since strategy switching may cause unintended security risks compromising the achieved protection, our solution includes mechanisms for detecting and mitigating this problem. The efficiency of our framework is evaluated experimentally.

We design a new efficient strategy synthesis method applicable to adversarial patrolling problems on graphs with arbitrary-length edges and possibly imperfect intrusion detection. The core ingredient is an efficient algorithm for computing the value and the gradient of a function assigning to every strategy its “protection” achieved. This allows for designing an efficient strategy improvement algorithm by differentiable programming and optimization techniques. Our method is the first one applicable to real-world patrolling graphs of reasonable sizes. It outperforms the state-of-the-art strategy synthesis algorithm by a margin.

We propose an algorithm for constructing efficient patrolling strategies in the Internet environment, where the protected targets are nodes connected to the network and the patrollers are software agents capable of detecting/preventing undesirable activities on the nodes. The algorithm is based on a novel compositional principle designed for a special class of strategies, and it can quickly construct (sub)optimal solutions even if the number of targets reaches hundreds of millions.

We consider the fixed-delay synthesis problem for continuous-time Markov chains extended with fixed-delay transitions (fdCTMC). The goal is to synthesize concrete values of the fixed-delays (timeouts) that minimize the expected total cost incurred before reaching a given set of target states. The same problem has been considered and solved in previous works by computing an optimal policy in a certain discrete-time Markov decision process (MDP) with a huge number of actions that correspond to suitably discretized values of the timeouts. In this paper, we design a symbolic fixed-delay synthesis algorithm which avoids the explicit construction of large action spaces. Instead, the algorithm computes a small sets of "promising" candidate actions on demand. The candidate actions are selected by minimizing a certain objective function by computing its symbolic derivative and extracting a univariate polynomial whose roots are precisely the points where the derivative takes zero value. Since roots of high degree univariate polynomials can be isolated very efficiently using modern mathematical software, we achieve not only drastic memory savings but also speedup by three orders of magnitude compared to the previous methods.

In this poster paper, we study performance of systems modeled by deterministic and stochastic Petri nets (DSPN). As a performance measure, we consider long-run average time spent in a set of markings. Even though this measure often appears in DSPN literature, its existence has never been considered. We provide a DSPN model of a simple communication protocol in which the long-run average time spent in a fixed marking is not well-defined due to a highly unstable behavior of the model. Further, we introduce a syntactical restriction on DSPN which preserves most of the modeling power yet guarantees existence of the long-run average.

Interactive Markov chains (IMC) are compositional behavioral models extending both labeled transition systems and continuous-time Markov chains. IMC pair modeling convenience - owed to compositionality properties - with effective verification algorithms and tools - owed to Markov properties. Thus far however, IMC verification did not consider compositionality properties, but considered closed systems. This paper discusses the evaluation of IMC in an open and thus compositional interpretation. For this we embed the IMC into a game that is played with the environment. We devise algorithms that enable us to derive bounds on reachability probabilities that are assured to hold in any composition context.

We introduce a new fragment of linear temporal logic (LTL) called LIO and a new class of Büchi automata (BA) called almost linear Büchi automata (ALBA). We provide effective translations between LIO and ALBA showing that the two formalisms are expressively equivalent. As we expect there to be applications of our results in model checking, we use two standard sources of specification formulae, namely Spec Patterns and BEEM, to study the practical relevance of the LIO fragment, and to compare our translation of LIO to ALBA with two standard translations of LTL to BA using alternating automata. Finally, we demonstrate that the LIO to ALBA translation can be much faster than the standard translation, and the resulting automata can be substantially smaller.

We propose deterministic timed automata (DTA) as a model-independent language for specifying performance and dependability measures over continuous-time stochastic processes. Technically, these measures are defined as limit frequencies of locations (control states) of a DTA that observes computations of a given stochastic process. Then, we study the properties of DTA measures over semi-Markov processes in greater detail. We show that DTA measures over semi-Markov processes are well-defined with probability one, and there are only finitely many values that can be assumed by these measures with positive probability. We also give an algorithm which approximates these values and the associated probabilities up to an arbitrarily small given precision. Thus, we obtain a general and effective framework for analysing DTA measures over semi-Markov processes.