EN FR
EN FR
DEVINE - 2025

2025Activity reportProject-Team​​​‌DEVINE

RNSR: 202424472A
  • Research‌ center Inria Centre at‌​‌ Rennes University
  • In partnership​​ with:CNRS, Université de​​​‌ Rennes
  • Team name: DEpendable‌ distributed systems: formal VerificatIoN‌​‌ made Efficient
  • In collaboration​​ with:Institut de recherche​​​‌ en informatique et systèmes‌ aléatoires (IRISA)

Creation of‌​‌ the Project-Team: 2024 January​​ 01

Each year, Inria​​​‌ research teams publish an‌ Activity Report presenting their‌​‌ work and results over​​ the reporting period. These​​​‌ reports follow a common‌ structure, with some optional‌​‌ sections depending on the​​ specific team. They typically​​​‌ begin by outlining the‌ overall objectives and research‌​‌ programme, including the main​​ research themes, goals, and​​​‌ methodological approaches. They also‌ describe the application domains‌​‌ targeted by the team,​​​‌ highlighting the scientific or​ societal contexts in which​‌ their work is situated.​​

The reports then present​​​‌ the highlights of the​ year, covering major scientific​‌ achievements, software developments, or​​ teaching contributions. When relevant,​​​‌ they include sections on​ software, platforms, and open​‌ data, detailing the tools​​ developed and how they​​​‌ are shared. A substantial​ part is dedicated to​‌ new results, where scientific​​ contributions are described in​​​‌ detail, often with subsections​ specifying participants and associated​‌ keywords.

Finally, the Activity​​ Report addresses funding, contracts,​​​‌ partnerships, and collaborations at​ various levels, from industrial​‌ agreements to international cooperations.​​ It also covers dissemination​​​‌ and teaching activities, such​ as participation in scientific​‌ events, outreach, and supervision.​​ The document concludes with​​​‌ a presentation of scientific​ production, including major publications​‌ and those produced during​​ the year.

Keywords

Computer​​​‌ Science and Digital Science​

  • A2.5.5. Software testing
  • A4.5.​‌ Formal method for verification,​​ reliability, certification
  • A4.5.2. Model-checking​​​‌
  • A8.9. Performance evaluation
  • A8.11.​ Game Theory

Other Research​‌ Topics and Application Domains​​

  • B5.1. Factory of the​​​‌ future
  • B6.6. Embedded systems​
  • B7.1. Traffic management

1​‌ Team members, visitors, external​​ collaborators

Research Scientists

  • Nathalie​​​‌ Bertrand [Team leader​, INRIA, Senior​‌ Researcher, HDR]​​
  • Loïc Hélouët [INRIA​​​‌, Senior Researcher,​ HDR]
  • Thierry Jéron​‌ [INRIA, Senior​​ Researcher, HDR]​​​‌

Faculty Members

  • Loïc Germerie-Guizouarn​ [UNIV RENNES,​‌ Associate Professor]
  • Aline​​ Goeminne [ENS RENNES​​​‌, Associate Professor,​ from Sep 2025]​‌
  • Julie Parreaux [UNIV​​ RENNES, Associate Professor​​​‌]

Post-Doctoral Fellows

  • Sayan​ Mukherjee [INRIA then​‌ CNRS, Post-Doctoral Fellow​​, from Feb 2025​​​‌]
  • Gaëtan Staquet [​INRIA, Post-Doctoral Fellow​‌, until Aug 2025​​]

PhD Students

  • Aymeric​​​‌ Come [INRIA]​
  • Victorien Desbois [NewLogUP​‌, CIFRE]
  • Pranav​​ Ghorpade [UNIV. SYDNEY​​​‌, located in Sydney​]
  • Luc Lapointe [​‌ENS PARIS-SACLAY, located​​ at LMF PARIS-SACLAY]​​​‌
  • Mathieu Laurent [ENS​ RENNES, shared with​‌ Magellan]
  • Luca Paparazzo​​ [UNIV RENNES]​​​‌
  • Antoine Thébault [Alstom​ , CIFRE, until​‌ Aug 2025]

Interns​​ and Apprentices

  • Maëlle Gautrin​​​‌ [ENS PARIS-SACLAY,​ Intern, from Mar​‌ 2025 until Aug 2025​​]
  • Nowar Kazem [​​​‌ENS RENNES, Intern​, from May 2025​‌ until Jul 2025]​​

Administrative Assistant

  • Laurence Dinh​​​‌ [INRIA]

Visiting​ Scientist

  • Srinivas Pinisetty [​‌IIT BHUBANESWAR, from​​ Jun 2025 until Jun​​​‌ 2025]

External Collaborator​

  • Ulrich Fahrenberg [EPITA​‌ then Université PARIS-SACLAY (since​​ Sept. 2025)]

2​​​‌ Overall objectives

Modern computer​ systems exploit concurrency in​‌ order to reach high​​ levels of performance. Moreover,​​​‌ an increasing number of​ complex applications are also​‌ distributed, either to attempt​​ to distribute the computation​​​‌ in search of performance​ gains, or because the​‌ system is by nature​​ made of several components​​​‌ that communicate over a​ network. Both aspects bring​‌ difficulties in the development​​ of dependable distributed applications.​​​‌ In fact, a large​ number of threads and​‌ components can mean that​​ the set of possible​​ behaviors of the system​​​‌ (thus, its set of‌ configurations) is prohibitively large.‌​‌ Beyond software, many physical​​ systems such as transportation​​​‌ networks or power grids‌ are massively distributed by‌​‌ nature. There also, the​​ variability of behaviors is​​​‌ extreme due to events‌ interleaving and unexpected faults.‌​‌ In the absence of​​ specific techniques for design​​​‌ and verification, ensuring the‌ functional correctness of such‌​‌ systems (i.e. their​​ ability to produce the​​​‌ expected outputs on given‌ inputs) can become extremely‌​‌ difficult.

Apart from their​​ distributed nature, many modern​​​‌ computers or physical systems‌ rely on extra-functional aspects‌​‌, such as real-time​​ constraints and stochasticity. Expressing​​​‌ the correctness of many‌ critical systems requires not‌​‌ only functional correctness, but​​ also several more complex​​​‌ properties related to execution‌ time, respect of deadlines,‌​‌ average performance, and probabilistic​​ convergence. We will more​​​‌ generally call these aspects‌ quantitative because they often‌​‌ involve enriching the description​​ of the systems, their​​​‌ requirements, and their analysis‌ with quantities such as‌​‌ time, probabilities, or other​​ measures such as cost.​​​‌ As illustrative examples, we‌ detail three applications in‌​‌ which a quantitative analysis​​ is crucial:

  • Time synchronization​​​‌ protocols are at the‌ heart of highly distributed‌​‌ computing systems such as​​ audio streaming over Ethernet,​​​‌ wireless sensor networks, accurate‌ distance measurement through GPS‌​‌ satellites, and cloud-based applications.​​ They must be fault-tolerant​​​‌ and should provide a‌ time measure with high‌​‌ accuracy in order for​​ these applications to be​​​‌ used, despite failures,latencies, etc.‌
  • In distributed computing, randomization‌​‌ can yield more efficient​​ solutions, or even permit​​​‌ solving problems that are‌ otherwise unsolvable, such as‌​‌ the consensus problem in​​ asynchronous message-passing systems in​​​‌ which as few as‌ one process can crash.‌​‌
  • Statistics on unpredictable events​​ are commonly used to​​​‌ enable performance evaluation, as‌ in regulation for transportation‌​‌ networks where journey duration​​ is represented as a​​​‌ random variable and potential‌ failures as random events.‌​‌

These three frameworks have​​ in common the fact​​​‌ that purely functional verification‌ techniques are certainly not‌​‌ sufficient for analyzing such​​ applications. Yet reasoning about​​​‌ quantitative distributed systems is‌ inherently difficult. For instance,‌​‌ the combination of distributed​​ aspects and probabilities makes​​​‌ human reasoning difficult; quoting‌ Lehmann and Rabin 63‌​‌: “proofs of correctness​​ for probabilistic distributed systems​​​‌ are extremely slippery”. Real-time‌ constraints bring additional difficulties‌​‌ since one not only​​ must reason about the​​​‌ computations but also their‌ timings which can induce‌​‌ particular interleavings between processes​​ that can be difficult​​​‌ to predict and debug‌ manually. Unfortunately, standard model-based‌​‌ verification techniques face scalability​​ issues on the large-scale​​​‌ applications we target, such‌ as transportation networks or‌​‌ time synchronization protocols in​​ large networks.

Yet, when​​​‌ the considered distributed systems‌ become bottlenecks in critical‌​‌ infrastructures, proving the correctness​​ or assessing the performance​​​‌ of such systems in‌ a reliable way is‌​‌ crucial. Indeed, failures can​​ be prohibitive in terms​​​‌ of financial cost or‌ even human loss. Safety-critical‌​‌ software or physical systems​​ that operate safely and​​​‌ dependably yield a competitive‌ advantage for industrials and‌​‌ reduce threats on the​​​‌ society. There is a​ need for the development​‌ of powerful algorithmic techniques​​ for providing a rich​​​‌ set of guarantees on​ such distributed systems, by​‌ ensuring their functional correctness,​​ while taking quantitative aspects​​​‌ into account. Our rationale​ is that quantitative aspects​‌ must be fully integrated​​ when reasoning about such​​​‌ systems, and must be​ at the heart of​‌ all phases of design,​​ requirement development, testing, bug​​​‌ finding, formal verification and​ synthesis. DEVINE aims to​‌ develop algorithms for ensuring​​ the dependability of quantitative​​​‌ distributed systems. To​ achieve this objective, we​‌ will develop model-based1​​ scalable verification and optimization​​​‌ techniques.

In order​ to ensure dependability of​‌ distributed systems with quantitative​​ aspects, the research agenda​​​‌ of DEVINE is structured​ into the following axes.​‌

  1. Efficient analysis of real-time​​ systems. The applicability of​​​‌ model-based formal methods to​ industrial-size real-time systems is​‌ challenged by the mix​​ in models of discrete​​​‌ and continuous variables. Efficient​ model-checking, testing and runtime​‌ verification algorithms are needed​​ to handle large models.​​​‌ We will also handle​ timing imprecisions and real-time​‌ security properties.
  2. Verification of​​ distributed algorithms. The behaviour​​​‌ of distributed algorithms and​ their implementations is hard​‌ to analyze due to​​ asynchrony and failures. We​​​‌ will develop innovative bug-finding​ techniques for MPI programs​‌ and verification methodologies for​​ pseudo-codes to prove their​​​‌ correctness independently of the​ number of processes.
  3. Optimization​‌ of multi-agent systems. Standard​​ optimization techniques do not​​​‌ scale to large multi-agent​ systems. We propose to​‌ formalize optimality, design efficient​​ planning algorithms, and explore​​​‌ the trade-off between strategy​ optimality and computation cost.​‌

A strength of the​​ model-based techniques and tools​​​‌ we develop is that​ they are generic and​‌ high-level, so that they​​ may prove useful in​​​‌ many application domains. The​ members of DEVINE aim​‌ at maintaining and increasing​​ strong relations with industrial​​​‌ partners. As for software,​ on the one hand,​‌ we will develop prototype​​ implementations to demonstrate the​​​‌ applicability of our techniques,​ and on the other​‌ hand, we will co-develop​​ specific tools answering the​​​‌ needs of industrials.

3​ Research program

3.1 Efficient​‌ analysis of real-time systems​​

Timed automata have been​​​‌ introduced in the early​ 1990s as a convenient​‌ framework for modelling and​​ reasoning about real-time systems​​​‌ 27. They combine​ discrete state space, to​‌ represent valuations of internal​​ variables, and continuous variables​​​‌ called clocks, e.g.​ to measure delays between​‌ events. Timed automata and​​ their variants have been​​​‌ extensively studied over the​ last 30 years, both​‌ on the theoretical and​​ practical sides. Several efficient​​​‌ tools have been developed​ and applied to industrial​‌ case studies 31,​​ 58. The efficiency​​​‌ of these tools is​ however still challenged by​‌ the mix of discrete​​ and continuous variables, which​​​‌ makes it hard to​ handle the state space​‌ symbolically. Our aim is​​ to develop techniques, algorithms,​​​‌ and tools that scale​ to larger models, which​‌ would allow us to​​ handle larger case studies.​​​‌

Efficient model checking algorithms​

As in many applications​‌ of formal methods, formal​​ verification suffers from state-space​​ explosion, which limits the​​​‌ scalability of algorithms unless‌ proper state-space reduction techniques‌​‌ are applied. In timed​​ automata, state-space explosion can​​​‌ be caused by two‌ factors: 1) a large‌​‌ discrete state space, e.g.​​ if the system is​​​‌ composed of many subsystems,‌ or contains several discrete‌​‌ variables; 2) a large​​ number of clocks or​​​‌ complex timing constraints. While‌ the first factor already‌​‌ appears in finite-state model​​ checking, in timed automata,​​​‌ the state space can‌ grow exponentially in the‌​‌ number of clocks requiring​​ a particular care. State-of-the-art​​​‌ algorithms can efficiently deal‌ with complex time constraints‌​‌ 54, 62,​​ 65 but fail at​​​‌ analyzing models with both‌ large discrete state spaces‌​‌ (for instance real-time distributed​​ systems) and real-time constraints.​​​‌ This is however crucial‌ to demonstrate the benefits‌​‌ of formal methods for​​ real-time systems on realistic​​​‌ applications.

We will build‌ novel tools based on‌​‌ compositional reasoning and predicate​​ abstraction to achieve formal​​​‌ verification performance comparable with‌ that of finite-state systems‌​‌ without explicit time constraints.​​ In fact, although clock​​​‌ constraints do cause state-space‌ explosion, they are often‌​‌ not the only source​​ of complexity; so smart​​​‌ ways of handling them‌ must be developed to‌​‌ handle large models. Ideally,​​ one should be able​​​‌ to handle clock variables‌ like any other variable‌​‌ in a program under​​ verification. In this context,​​​‌ predicate abstraction 53,‌ 69 is a promising‌​‌ direction since when used​​ properly, clock variables can​​​‌ be treated as any‌ other system variable; so‌​‌ several techniques from software​​ model checking or finite-state​​​‌ automata can be applied.‌ Compositionality is a well-known‌​‌ approach to handle larger​​ models 41; we​​​‌ will develop techniques to‌ specifically handle the timing‌​‌ aspects in such approaches,​​ and target the development​​​‌ of both fully automatized‌ and interactive compositional model‌​‌ checkers. Last, some performance​​ achievements might appear by​​​‌ targeting specific applications and‌ developing tailored algorithms, rather‌​‌ than relying on one​​ generic algorithm. Our collaborations​​​‌ with industrial partners will‌ guide us in this‌​‌ direction since these are​​ opportunities to consider specific​​​‌ practical problems. In all‌ these works, we will‌​‌ use and contribute to​​ the open-source timed automata​​​‌ model checker TChecker 58‌.

Testing and runtime‌​‌ verification

To extend the​​ applicability of models like​​​‌ timed automata to verify‌ industrial-size real-time systems, one‌​‌ can relax the exhaustiveness​​ guarantee provided by model​​​‌ checking. Model-based test synthesis‌ is one such technique‌​‌ (see, e.g. 73):​​ it consists in synthesizing,​​​‌ from a system model,‌ sequences of actions to‌​‌ be performed on the​​ implementation, in order to​​​‌ check that it behaves‌ as specified. We will‌​‌ generate such test cases​​ from real-time requirements, leveraging​​​‌ techniques recently developed by‌ team members based on‌​‌ both test synthesis from​​ game theory 57 and​​​‌ consistency checking 61;‌ this will complement the‌​‌ tool suite we developed​​ in our collaboration with​​​‌ MERCE for checking consistency‌ of real-time requirements, to‌​‌ obtain test cases for​​ checking those requirements on​​​‌ real implementations.

Runtime monitoring‌ 30 is another verification‌​‌ technique for assessing the​​​‌ validity of properties at​ runtime: it consists in​‌ observing the system as​​ it executes and deciding​​​‌ as soon as possible​ whether the properties are​‌ satisfied or violated. In​​ many contexts, the system​​​‌ is only partially observable,​ i.e. some internal actions​‌ are hidden to the​​ monitor. Runtime monitoring is​​​‌ however limited to real-time​ systems that are not​‌ distributed. Our objective is​​ to develop a framework​​​‌ for distributed runtime monitoring​ of real-time systems, in​‌ which several monitors observe​​ components of the system,​​​‌ and exchange information so​ as to decide as​‌ early as possible on​​ the validity of the​​​‌ property. Efficient solutions should​ limit the amount of​‌ communication as well as​​ the computation time. Timed​​​‌ markings, a formalism we​ introduced and recently used​‌ to efficiently compute and​​ manipulate sets of configurations​​​‌ 37, are a​ natural candidate tool to​‌ use. In the distributed​​ setting, however, timed markings​​​‌ need to be reshaped​ to store sufficient information,​‌ and also to enable​​ efficient updates with the​​​‌ observation and information stream.​

Timing imprecisions

The model​‌ of timed automata for​​ real-time systems assumes arbitrary​​​‌ precision in time measurements.​ This artifact which is​‌ theoretically convenient has the​​ drawback of missing behaviours​​​‌ if delays are slightly​ shifted. Since time drifts​‌ are inevitable in distributed​​ systems, we will pursue​​​‌ the development of models,​ semantics and algorithms to​‌ take timing imprecisions into​​ account for real-time systems.​​​‌ For instance, the efficiency​ of our recent algorithm​‌ for synthesizing permissive strategies​​ 44 can be improved​​​‌ by relaxing its precision​ while keeping track of​‌ the amount of approximation​​ in the computation.

Timing​​​‌ imprecisions are also very​ relevant in online techniques,​‌ such as monitoring, testing​​ and learning, since those​​​‌ techniques involve interactions with​ physical implementations. Because of​‌ those imprecisions, the observation​​ of the system may​​​‌ be inexact, and the​ actions performed on the​‌ system may be slightly​​ shifted, so that a​​​‌ given sequence of inputs​ may result in different​‌ outputs. This does not​​ fit with our current​​​‌ techniques 56, 57​, and we will​‌ have to develop specific​​ approaches to take such​​​‌ imprecisions into account.

Real-time​ security properties

Most often​‌ in formal methods properties​​ are defined at the​​​‌ level of individual behaviours:​ for instance, an execution​‌ of a program is​​ terminating, or it isn't.​​​‌ However, in order to​ express security properties such​‌ as non-interference, one needs​​ to reason on pairs​​​‌ of executions (for instance​ to guarantee that observing​‌ the control flow of​​ a program does not​​​‌ leak information on a​ private key), or more​‌ generally sets of executions.​​ So-called hyperproperties, introduced​​​‌ a decade ago 43​, 42, allow​‌ one to compare executions​​ of untimed models. Dealing​​​‌ with real-time in that​ context is a great​‌ challenge since one needs​​ to compare dates of​​​‌ event occurrences. So far​ it only has been​‌ considered in a discrete-time​​ setting 36. We​​​‌ will provide verification algorithms​ in the continuous-time case,​‌ with the objective of​​ addressing security properties related​​ to timing issues, such​​​‌ as covert communication induced‌ by timing channels.

3.2‌​‌ Verification of distributed algorithms​​

Distributed algorithms are central​​​‌ to many domains such‌ as scientific computing, telecommunications‌​‌ and the blockchain. Even​​ when they aim at​​​‌ performing simple tasks, their‌ behaviour is hard to‌​‌ analyze, due mainly to​​ the asynchrony between the​​​‌ processes and to the‌ presence of faults (crashes,‌​‌ message losses, etc.). We​​ aim on the one​​​‌ hand at designing efficient‌ techniques for verification of‌​‌ HPC (High Performance Computing)​​ programs, and on the​​​‌ other hand at establishing‌ correctness of distributed algorithms‌​‌ independently of the number​​ of participants.

Efficient bug​​​‌ finding for MPI program‌

HPC applications in the‌​‌ MPI (message passing interface)​​ programming model consist of​​​‌ distributed programs that communicate‌ asynchronously through FIFO (First‌​‌ In First Out) channels.​​ Finding bugs, or proving​​​‌ their absence, in such‌ programs is challenging because‌​‌ of the complexity due​​ to concurrency and communication.​​​‌ The goal of the‌ McSimGrid tool 66 is‌​‌ to automatically check properties​​ directly on the programs,​​​‌ considering all alternative executions‌ of the program for‌​‌ a fixed input. Rather​​ than proving correctness, it​​​‌ aims at finding concurrency‌ and communication bugs efficiently.‌​‌ As often in verification,​​ scaling to real programs​​​‌ requires techniques that avoid‌ the state-space explosion. One‌​‌ way to do so,​​ is to use dynamic​​​‌ partial order reduction (DPOR)‌ 52, 29,‌​‌ 68, which cleverly​​ exploits the independence of​​​‌ concurrent events to reduce‌ the state space to‌​‌ be explored. Beyond plain​​ DPOR algorithms, in order​​​‌ to go an order‌ of magnitude further in‌​‌ terms of program size,​​ we propose to combine​​​‌ DPOR with other efficient‌ bug-finding techniques, such as‌​‌ directed model checking 50​​. Directed model checking​​​‌ prioritizes the state-space exploration‌ using A*-like‌​‌ algorithms (that is optimal​​ pathfinding algorithms), relying on​​​‌ approximate distances to a‌ goal. In the context‌​‌ of MPI programs, the​​ definition of appropriate distances​​​‌ is crucial for balancing‌ the trade-off between precision‌​‌ and computation cost, that​​ impact the time to​​​‌ find an error in‌ two opposite ways. Alternatively,‌​‌ one can combine DPOR​​ with other bug-finding techniques,​​​‌ by bounding some values‌ 45 or progressively refining‌​‌ the independence relation. In​​ order to evaluate these​​​‌ heuristics for MPI programs,‌ the various approaches will‌​‌ be implemented in McSimGrid​​ and benchmarked against academic​​​‌ case studies.

Parameterized verification‌ of pseudo-codes

The correctness‌​‌ of distributed algorithms should​​ be established independently of​​​‌ actual setup, i.e. the‌ number of processes, the‌​‌ potential failures, and the​​ communication topology when relevant.​​​‌ Parameterized verification answers this‌ need by handling multiple‌​‌ model-checking queries at once;​​ it also often comes​​​‌ with cutoff results that‌ provide bounds on the‌​‌ parameters for which a​​ bug can happen, in​​​‌ case the correctness does‌ not hold. To overcome‌​‌ the general undecidability of​​ the verification of properties​​​‌ for distributed systems with‌ an unbounded number of‌​‌ processes 28, we​​ propose two natural approaches:​​​‌ on the one hand‌ exhibiting models for specific‌​‌ distributed algorithms with decidable​​​‌ parameterized verification, and on​ the other hand developing​‌ incomplete verification algorithms that​​ may not terminate or​​​‌ may be inconclusive on​ some instances. We illustrate​‌ these alternatives on the​​ two archetypal frameworks below,​​​‌ on which we will​ focus first.

Parameterized verification​‌ of models of MPI​​ programs. Complementary to the​​​‌ objective of efficiently finding​ bugs in MPI programs,​‌ one can come up​​ with models of MPI​​​‌ programs and study their​ parameterized verification problem. A​‌ relevant way to classify​​ MPI programs with respect​​​‌ to model checking is​ by the communication primitives​‌ they use and possibly​​ the communication topologies: mailboxes​​​‌ are typically represented as​ queues or bags, for​‌ instance. Our objective is​​ to define classes of​​​‌ models for MPI programs​ with decidable parameterized verification.​‌ Most likely these models​​ will approximate the actual​​​‌ behaviour of the programs.​ A prototype implementing parameterized​‌ verification for MPI programs​​ would be a major​​​‌ breakthrough compared to the​ fixed-instances existing tools such​‌ as CIVL 64.​​

Parameterized verification of randomized​​​‌ distributed algorithms. Randomization is​ an elegant tool to​‌ design efficient algorithms or​​ even to solve problems​​​‌ otherwise unsolvable, especially in​ distributed computing, where probabilities​‌ break symmetry between the​​ components. Till now, automated​​​‌ proofs of randomized distributed​ algorithms remain limited to​‌ restricted types of algorithms,​​ restricted classes of schedulers,​​​‌ and restricted properties 32​, 33, 34​‌. Leveraging parameterized verification​​ techniques to handle randomized​​​‌ distributed algorithms raises the​ challenge that performance typically​‌ depend on the number​​ of participants, whereas standard​​​‌ parameterized verification techniques abstract​ away this parameter. Counter-example​‌ guided abstraction refinement (CEGAR)​​ approaches have proven extremely​​​‌ efficient on non-randomized fault-tolerant​ distributed algorithms 35,​‌ 70. Building on​​ the latter, and on​​​‌ an existing CEGAR framework​ for finite-state probabilistic systems​‌ 59, 48,​​ we aim at targetting​​​‌ randomized distributed algorithms, which​ is non-trivial: appropriate predicates​‌ must be defined, and​​ counterexamples must be generic​​​‌ enough to account for​ sets of parameter values.​‌

3.3 Optimization of multi-agent​​ systems

While verification merely​​​‌ checks that a property​ holds, control aims at​‌ optimizing the system performance​​ related to quantities such​​​‌ as time or energy​ consumption. A controller resolves​‌ choices that are left​​ open, typically during early​​​‌ design phases to adapt​ to a particular context.​‌ Resolving choices amounts to​​ selecting a strategy, in​​​‌ order to optimize certain​ quantitative objectives. When choices​‌ are left to several​​ agents, their interaction is​​​‌ represented by a game.​ Solving a game amounts​‌ to computing optimal strategies​​ for each of the​​​‌ agents. Controlling large systems​ or solving large games​‌ is even more challenging​​ than verification: the distributed​​​‌ nature of multi-agent systems​ leads to exponential blowup​‌ of the state space.​​ Admittedly, standard optimization techniques​​​‌ 67 such as value​ iteration or policy iteration​‌ (that is, iterative fixpoint​​ computations of optimal strategies)​​​‌ do not scale to​ large state spaces. Motivated​‌ by traffic management in​​ transportation systems, we will​​​‌ attack this problem by​ formalising optimality logically, designing​‌ efficient planning algorithms or​​ computing close-to-optimal strategies with​​ guarantees.

Formalising optimality in​​​‌ quantitative games

Before even‌ computing optimal or quasi-optimal‌​‌ strategies for multi-agent systems,​​ one needs to formally​​​‌ define these. Strategy Logic‌ (SL  for short)‌​‌ is a temporal logic​​ for expressing complex properties​​​‌ of multiple-player games 40‌. It can be‌​‌ used to express, e.g.​​ the existence of equilibria,​​​‌ as well as properties‌ of the outcomes of‌​‌ those equilibria. However, this​​ logic is not suited​​​‌ to games with quantitative‌ aspects: it cannot handle‌​‌ quantitative (as opposed to​​ Boolean) payoffs for the​​​‌ players, which are needed‌ for fine-grained representation of‌​‌ their preferences. We will​​ continue the exploration of​​​‌ such quantitative features of‌ SL, building on‌​‌ our recent works 38​​ on fuzzy SL (which​​​‌ was, to our knowledge,‌ the first decidable quantitative‌​‌ extension of SL).​​ Due to the high​​​‌ complexity of model checking‌ for SL, we‌​‌ will also consider fragments​​ of the logic, targeting​​​‌ good trade-offs between expressiveness‌ and complexity.

Computing sub-optimal‌​‌ strategies with guarantees

In​​ non-critical contexts, sub-optimal strategies​​​‌ can be sufficient, and‌ their computation can be‌​‌ less costly than optimal​​ ones. In game theory,​​​‌ Simon 71 introduced the‌ portmanteau word satisficing that‌​‌ combines satisfying and sufficient​​, to describe situations​​​‌ in non-cooperative games, where‌ agents may opt for‌​‌ a non-optimal strategy if​​ they think their reward​​​‌ is good enough and‌ that the effort (be‌​‌ it time, energy, or​​ computational power) needed to​​​‌ get closer to the‌ optimal reward would be‌​‌ prohibitive. This notion has​​ been addressed in economy​​​‌ as an alternative to‌ optimality in standard reinforcement‌​‌ learning (RL) algorithms (see,​​ e.g. 72). Strategies​​​‌ in RL are built‌ to optimize quantitative goals‌​‌ but allow stopping construction​​ when side measures such​​​‌ as regrets or risk‌ exceed a fixed aspiration‌​‌ level. Building on our​​ expertise in various aspects​​​‌ of quantitative games 39‌, 55, we‌​‌ aim at computing good-enough​​ strategies by relating the​​​‌ improvement, e.g. measured as‌ a reward, that can‌​‌ be obtained by changing​​ a strategy and the​​​‌ extra effort needed to‌ play the new strategy.‌​‌

We will also address​​ the scalability issue in​​​‌ optimal control problems in‌ Markov decision processes by‌​‌ targeting distributed systems. This​​ direction will combine approaches​​​‌ both from the formal‌ verification world, such as‌​‌ abstraction techniques, and bounded​​ verification 49, and​​​‌ from the reinforcement learning‌ world, such as multi-agent‌​‌ reinforcement learning techniques and​​ approximate solution methods (e.g.​​​‌ deep learning). This rich‌ set of techniques will‌​‌ allow us to solve​​ applications such as train​​​‌ regulation problems.

Application to‌ traffic management in transports‌​‌

Urban rail transportation calls​​ for optimization techniques in​​​‌ order to improve the‌ users experience in terms‌​‌ of metro punctuality, regularity,​​ and time needed to​​​‌ resume to a nominal‌ behaviour after an incident,‌​‌ to name a few.​​ For a single metro​​​‌ line, efficient traffic management‌ techniques exist and mainly‌​‌ consist in adapting fleets​​ sizes, speeds of trains​​​‌ and dwell times 46‌. However, in large‌​‌ cities, the public transport​​​‌ demand is expected to​ grow tremendously (49% larger​‌ in 2050 than in​​ 2012) with huge economic​​​‌ and environmental impacts. Improving​ efficiency of public transports​‌ must include several transport​​ modes and several operators,​​​‌ making the state-of-the-art techniques​ unapplicable. Multi-modal transport raises​‌ the issue that operators,​​ clients, and decision-makers have​​​‌ their own objective. Optimizing​ traffic management in this​‌ context is a challenging​​ task: it amounts to​​​‌ solving a huge multi-player​ game with multiple objectives.​‌ Our aim is to​​ compute good-enough strategies for​​​‌ each actor in these​ games. Despite the large​‌ size of transport networks,​​ recently developed tools for​​​‌ concurrent hybrid models enable​ numerical methods, and can​‌ be used to learn​​ regulation mechanisms 51.​​​‌ Symbolic representations of sets​ of states and distributions​‌ 60 is also promising​​ to speed up simulation,​​​‌ discover appropriate abstractions and​ therefore enable efficient traffic​‌ management.

4 Application domains​​

A strength of the​​​‌ model-based techniques and tools​ we develop is to​‌ be generic and high-level​​ so that they may​​​‌ find applications in many​ domains. Members of the​‌ team already have long-lasting​​ collaborations with industrial partners​​​‌ in transportations systems and​ factory automation, and new​‌ collaborations in blockchain technologies​​ are emerging. Our experience​​​‌ demonstrates that new applications​ often feed our research​‌ with new and challenging​​ problems. We are however​​​‌ aware of the time​ required to invest into​‌ a new domain. Our​​ strategy is hence to​​​‌ be opportunistic and remain​ open to any potential​‌ application of our techniques,​​ within the limits imposed​​​‌ by the size of​ the team.

Industrial transfer​‌

In terms of industrial​​ technology transfer, we will​​​‌ aim in DEVINE at​ maintaining and increasing strong​‌ relations with industrial partners​​ from various application domains.​​​‌ The genericity of formal​ methods and verification techniques​‌ indeed open many opportunities​​ for industrial transfer. We​​​‌ aim at transferring knowledge​ to researchers and engineers​‌ in industrial partners through​​ CIFRE PhD projects, and​​​‌ impact the products of​ our industrial partners with​‌ innovative techniques that can​​ lead to patents and​​​‌ be used in production.​

These objectives are very​‌ sensible given the long-term​​ experience of several members​​​‌ of DEVINE in collaborations​ with Mitsubishi Electric (MERCE)​‌ and Alstom Transport. These​​ have led to several​​​‌ CIFRE PhDs, to training​ of an engineer from​‌ MERCE, to filing of​​ patents, publications, and to​​​‌ the transfer of a​ software tool (SIMSTORS) to​‌ Alstom Transport.

Transfer to​​ other CS fields

Apart​​​‌ from transfer to industries,​ we will also aim​‌ at impacting fields in​​ computer science other than​​​‌ formal methods. The two​ fields we target are​‌ distributed computing and scientific​​ computing, which would in​​​‌ our opinion benefit from​ our expected contributions in​‌ formal verification of distributed​​ algorithms and verification of​​​‌ MPI programs. We plan​ to continue and strengthen​‌ our impact to these​​ fields, mainly through publications​​​‌ in conferences and journals​ of the respective domains,​‌ that members of DEVINE​​ started to do via​​​‌ collaborations with distributed computing​ and MPI experts.

5​‌ Social and environmental responsibility​​

5.1 Footprint of research​​ activities

Some members of​​​‌ the team individually take‌ part in the TCS4F‌​‌ (Theoretical Computer Science for​​ Future) initiatives.

Since​​​‌ COVID, the carbon footprint‌ of travels related to‌​‌ our research activities significantly​​ decreased. Some members of​​​‌ the team no longer‌ fly to attend conferences,‌​‌ and everyone carefully chooses​​ the conferences where they​​​‌ submit not only in‌ terms of reputation, but‌​‌ also taking into account​​ the location or the​​​‌ possibility to attend online.‌

5.2 Impact of research‌​‌ results

DEVINE team members​​ maintain long-term relationships with​​​‌ industrials. These tight collaborations‌ could have impact in‌​‌ the future, as illustrated​​ below by two examples.​​​‌

  • Since 2022, DEVINE collaborates‌ with Alstom Transport on‌​‌ improvement of traffic management​​ in metro systems (CIFRE​​​‌ grant of Antoine Thébault).‌ One of the concerns‌​‌ in this study is​​ to reduce the energy​​​‌ consumed by metros using‌ smart controllers. Though these‌​‌ studies are currently conducted​​ at a theoretical level​​​‌ and tested in silico,‌ their transfer to running‌​‌ systems may help reducing​​ the energy used in​​​‌ urban transports in the‌ future.
  • In 2025, DEVINE‌​‌ and other Inria teams​​ started a Défi commun​​​‌ with MERCE (Mitsubishi Electric‌ Research Center Europe). The‌​‌ outcomes of this 4-year​​ long project should impact​​​‌ the design of safe‌ systems in particular in‌​‌ several domains like railway,​​ automotive, FA, elevators, where​​​‌ Mitsubishi is a world-class‌ leader.

6 Highlights of‌​‌ the year

MERCE (Mitsubishi​​ Electric Research Center Europe)​​​‌ and Inria have launched‌ the Défi commun FRAIME‌​‌ mid-september 2025. In FRAIME,​​ we propose to explore​​​‌ on the one hand‌ how Formal Methods can‌​‌ provide guarantees on AI​​ systems, and on the​​​‌ other hand how AI‌ can help Formal Methods‌​‌ to be more efficient​​ and easier to use​​​‌ by practitioners. The vision‌ is to intertwine Formal‌​‌ Methods and AI to​​ efficiently design safe systems.​​​‌

7 Latest software developments,‌ platforms, open data

7.1‌​‌ Latest software developments

7.1.1​​ ParaGraphs

  • Name:
    ParaGraphs
  • Keywords:​​​‌
    Game theory, Symbolic verification‌
  • Functional Description:
    ParaGraphs implements‌​‌ two new algorithms based​​ on antichains, and the​​​‌ original PSPACE algorithm for‌ solving parameterized concurrent games‌​‌ with reachability objectives. It​​ is an open-source C++20​​​‌ tool. It currently supports‌ constraints given as finite‌​‌ unions of intervals, but​​ is designed in a​​​‌ modular way and permits‌ to easily define new‌​‌ types (for instance semi-linear​​ sets), as well as​​​‌ specific implementations of the‌ algorithms for these types.‌​‌
  • Release Contributions:
    First version​​ of the software.
  • URL:​​​‌
  • Contact:
    Nathalie Bertrand‌
  • Participants:
    Gaetan Staquet, Nathalie‌​‌ Bertrand

7.1.2 SimGrid

  • Keywords:​​
    Large-scale Emulators, Grid Computing,​​​‌ Distributed Applications
  • Scientific Description:‌

    SimGrid is a toolkit‌​‌ that provides core functionalities​​ for the simulation of​​​‌ distributed applications in heterogeneous‌ distributed environments. The simulation‌​‌ engine uses algorithmic and​​ implementation techniques toward the​​​‌ fast simulation of large‌ systems on a single‌​‌ machine. The models are​​ theoretically grounded and experimentally​​​‌ validated. The results are‌ reproducible, enabling better scientific‌​‌ practices.

    Its models of​​ networks, cpus and disks​​​‌ are adapted to (Data)Grids,‌ P2P, Clouds, Fogs, Clusters‌​‌ and HPC, allowing multi-domain​​​‌ studies. It can be​ used either to simulate​‌ algorithms and prototypes of​​ applications, or to emulate​​​‌ real MPI applications through​ the virtualization of their​‌ communication, or to formally​​ assess algorithms and applications​​​‌ that can run in​ the framework.

    The formal​‌ verification module explores all​​ possible message interleavings in​​​‌ the application, searching for​ states violating the provided​‌ properties. This tool can​​ be used to assess​​​‌ safety properties over arbitrary​ and legacy codes, thanks​‌ to a system-level introspection​​ tool that provides a​​​‌ finely detailed view of​ the running application to​‌ the model checker. This​​ can for example be​​​‌ leveraged to verify arbitrary​ MPI code written in​‌ C/C++/Fortran.

  • Functional Description:
    SimGrid​​ is a simulation toolkit​​​‌ that provides core functionalities​ for the simulation of​‌ distributed applications in large​​ scale heterogeneous distributed environments.​​​‌
  • Release Contributions:

    Breaking the​ seal: v4.0 was not​‌ the final release.

    *​​ Allow one to unseal​​​‌ netzones to modify the​ platform even after the​‌ simulation start. * The​​ model-checker can now report​​​‌ memory race conditions (see​ tutorial). * Pip builds​‌ should now work out​​ of the box. *​​​‌ (+ the usual bug​ fixes overall, and improvements​‌ to the Java/Python bindings).​​

  • News of the Year:​​​‌

    Most of the scientific​ work in SimGrid in​‌ 2025 occurred in the​​ embedded model checker. Our​​​‌ work could profit of​ the 2 major releases​‌ of SimGrid in 2025​​ (v4.0 in March and​​​‌ Release v4.1 in November).​ We first worked on​‌ improving the performance of​​ the verification process, and​​​‌ fixed some bugs. Our​ preferred algorithm (ODPOR reduction​‌ + Best-first exploration to​​ enable random walk) is​​​‌ now much faster and​ consumes much less memory,​‌ being 5x faster on​​ small scenarios. But since​​​‌ performance is now linear​ with the number of​‌ states, while it used​​ to be polynomial, 5x​​​‌ faster is the lowest​ performance boost you can​‌ expect from the new​​ version. The number of​​​‌ states to explore for​ a given scenario is​‌ still the same (ODPOR​​ was not improved), but​​​‌ we now can explore​ these states much faster.​‌

    In addition, we added​​ the ability to check​​​‌ for race conditions in​ the user code. This​‌ feature relies on a​​ specific pass to the​​​‌ clang LLVM compiler, to​ instrument the memory accesses.​‌ We use this feature​​ for teaching purposes at​​​‌ our institutions.

    We are​ working on a parallel​‌ explorer leveraging all cores​​ to accelerate the exploration,​​​‌ but unfortunately, we did​ not manage to find​‌ all the bugs in​​ our parallel explorer yet.​​​‌ We are working on​ verifying SimGrid with itself​‌ to hammer this bug​​ out.

  • URL:
  • Publication:​​​‌
  • Contact:
    Martin Quinson​
  • Participants:
    Mathieu Laurent, Anne-Cécile​‌ Orgerie, Arnaud Legrand, Augustin​​ Degomme, Arnaud Giersch, Frédéric​​​‌ Suter, Martin Quinson, Samuel​ Thibault
  • Partners:
    CNRS, ENS​‌ Rennes

7.1.3 Ticynet

  • Name:​​
    Time Cyclic Networks Tool​​​‌
  • Keywords:
    Simulator, Public transport,​ Transport model
  • Functional Description:​‌

    Ticynet is a tool​​ for the analysis of​​​‌ quantitative properties of urban​ transport networks and for​‌ the optimization of these​​ properties through control techniques.​​

    Specifically, Ticynet allows users​​​‌ to design a metro‌ network and analyze the‌​‌ energy savings achieved through​​ an appropriate fleet management​​​‌ strategy.

    In its current‌ version, Ticynet allows to:‌​‌

    - model an urban​​ transport line (tracks and​​​‌ vehicle fleet) - model‌ the regenerative braking phenomenon‌​‌ - integrate a controller​​ managing vehicle departures, arrivals,​​​‌ and speeds - simulate‌ the complete system to‌​‌ quantify the energy saved​​ ratio.

  • Release Contributions:
    Initial​​​‌ version, demonstrator for the‌ ANR Project BisoUS.
  • Contact:‌​‌
    Loic Helouet

8 New​​ results

This section presents​​​‌ the contributions of the‌ team members in 2025.‌​‌

8.1 New results on​​ efficient analysis of real-time​​​‌ systems

Participants: Ulrich Fahrenberg‌, Loïc Germerie-Guizouarn,‌​‌ Loïc Hélouët, Thierry​​ Jéron, Sayan Mukherjee​​​‌, Julie Parreaux,‌ Ocan Sankur, Gaëtan‌​‌ Staquet.

On this​​ research axis, our contributions​​​‌ this year concern: runtime‌ enforcement, model learning for‌​‌ real-time systems, timed games,​​ and compositionality of transducers.​​​‌

Runtime enforcement

In the‌ paper 14, we‌​‌ deal with the problem​​ of runtime enforcement (RE)​​​‌ in the context of‌ reactive systems, which consists‌​‌ in modifying the outputs​​ of a system minimally​​​‌ to ensure its correctness.‌ In contrast to enforcers‌​‌ that can postpone events​​ via buffering, enforcers for​​​‌ reactive systems must operate‌ within the same reactive‌​‌ cycle, always yielding a​​ (possibly modied) output. For​​​‌ safety properties, an enforcer‌ makes sure to satisfy‌​‌ the property at each​​ step. However, for general​​​‌ regular properties, one can‌ only expect to satisfy‌​‌ the property eventually. There​​ is then a risk​​​‌ that even under enforcement‌ the satisfaction is indefinitely‌​‌ delayed, and the property​​ is never actually satisfied.​​​‌ Forcing the satisfaction of‌ regular or ω-regular‌​‌ properties has been considered​​ using bounded fairness and​​​‌ prompt eventuality. In this‌ paper, we propose a‌​‌ new runtime enforcement framework​​ for regular properties with​​​‌ prompt eventualities. Given an‌ automaton specifying a property‌​‌ φ and a bound​​ k, the enforcer​​​‌ should never falsify φ‌ more than k consecutive‌​‌ steps. We formally define​​ this RE problem, characterize​​​‌ k-enforceability of automata,‌ and exhibit the construction‌​‌ of the enforcer. Rather​​ than fixing k,​​​‌ we also study whether‌ k can be computed‌​‌ to ensure k-enforceability​​ or some maximal coverage​​​‌ of φ. We‌ implement the k-prompt‌​‌ enforcement framework, and demonstrate​​ its behaviour with varying​​​‌ k.

Learning models‌ for real-time systems

In‌​‌ 22 we present a​​ state-merging algorithm for learning​​​‌ timed languages definable by‌ Event-Recording Automata (ERA) using‌​‌ positive and negative samples​​ in the form of​​​‌ symbolic timed words. Our‌ algorithm, (Learning Event-recording Automata‌​‌ Passively), constructs a possibly​​ nondeterministic ERA from such​​​‌ samples based on merging‌ techniques. We prove that‌​‌ determining whether two ERA​​ states can be merged​​​‌ while preserving sample consistency‌ is an NP-complete problem,‌​‌ and address this with​​ a practical SMT-based solution.​​​‌ Our implementation demonstrates the‌ algorithm’s effectiveness through examples.‌​‌ We also show that​​ every ERA-definable language can​​​‌ be inferred using our‌ algorithm with a suitable‌​‌ sample.

In 18 we​​​‌ present the first algorithm​ for query learning Mealy​‌ machines with timers in​​ a black-box context. Our​​​‌ algorithm is an extension​ of the L♯​‌ algorithm of Vaandrager et​​ al. to a timed​​​‌ setting. We rely on​ symbolic queries which empower​‌ us to reason on​​ untimed executions while learning.​​​‌ Similarly to the algorithm​ for learning timed automata​‌ of Waga, these symbolic​​ queries can be realized​​​‌ using finitely many concrete​ queries. Experiments with a​‌ prototype implementation show that​​ our algorithm is able​​​‌ to efficiently learn realistic​ benchmarks.

Timed Games

Weighted​‌ Timed Games (WTG for​​ short) are the most​​​‌ widely used model to​ describe controller synthesis problems​‌ involving real-time issues. We​​ consider optimal reachability objectives,​​​‌ in which one of​ the players, that we​‌ call Min, wants to​​ reach a target location​​​‌ while minimising the cumulated​ weight. Unfortunately, WTGs are​‌ notoriously difficult, and undecidable​​ with two or more​​​‌ clocks. As a consequence,​ one-clock WTGs have attracted​‌ a lot of attention,​​ especially because they are​​​‌ known to be decidable​ when only non-negative weights​‌ are allowed. However, when​​ arbitrary weights are considered,​​​‌ despite several recent works,​ their decidability status was​‌ still unknown. In 9​​, we solve this​​​‌ problem positively and show​ that the value function​‌ can be computed in​​ exponential time (if weights​​​‌ are encoded in unary).​ Alternatively to restricting to​‌ a single clock, several​​ conditions, one of them​​​‌ being divergence, have been​ given to recover decidability.​‌ In such weighted timed​​ games (like in untimed​​​‌ weighted games in the​ presence of negative weights),​‌ Min may need finite​​ memory to play (close​​​‌ to) optimally. This is​ thus tempting to try​‌ to emulate this finite​​ memory with other strategic​​​‌ capabilities. In 10,​ we allow the players​‌ to use stochastic decisions,​​ both in the choice​​​‌ of transitions and of​ timing delays. We give​‌ a definition of the​​ expected value in weighted​​​‌ timed games. We then​ show that, in divergent​‌ weighted timed games as​​ well as in (untimed)​​​‌ weighted games (that we​ call shortest-path games in​‌ the following), the stochastic​​ value is indeed equal​​​‌ to the classical (deterministic)​ value, thus proving that​‌ Min can guarantee the​​ same value while only​​​‌ using stochastic choices, and​ no memory.

Compositionality of​‌ transducers

Deterministic two-way transducers​​ with pebbles (aka pebble​​​‌ transducers) capture the class​ of polyregular functions, which​‌ extend the string-to-string regular​​ functions allowing polynomial growth​​​‌ instead of linear growth.​ They are the object​‌ of 20. One​​ of the most fundamental​​​‌ operations on functions is​ composition, and (poly)regular functions​‌ can be realized as​​ a composition of several​​​‌ simpler functions. In general,​ composition of deterministic two-way​‌ transducers incur a doubly​​ exponential blow-up in the​​​‌ size of the inputs.​ A major improvement in​‌ this direction comes from​​ the fundamental result of​​​‌ Dartois et al. 47​ showing a polynomial construction​‌ for the composition of​​ reversible two-way transducers. A​​​‌ precise complexity analysis for​ existing composition techniques of​‌ pebble transducers is missing,​​ but they rely on​​ the classic composition of​​​‌ two-way transducers and inherit‌ the double exponential complexity.‌​‌ To overcome this problem,​​ we introduce in this​​​‌ paper reversible pebble transducers.‌ Our main results are‌​‌ efficient uniformization techniques for​​ non-deterministic pebble transducers to​​​‌ reversible ones and efficient‌ composition for reversible pebble‌​‌ transducers. Our objective is​​ now to lift these​​​‌ result to the timed‌ setting.

8.2 New results‌​‌ on verification of distributed​​ algorithms

Participants: Nathalie Bertrand​​​‌, Pranav Ghorpade,‌ Thierry Jéron, Mathieu‌​‌ Laurent.

On this​​ research axis, our contributions​​​‌ this year range from‌ fundamental to more applied:‌​‌ true concurrency models such​​ as Petri nets, dynamical​​​‌ partial order reduction techniques,‌ and verification of blockchain‌​‌ protocols.

True concurrency models​​

In 13 we consider​​​‌ a new approach for‌ the concurrent semantics of‌​‌ Petri nets. Petri nets​​ and their variants are​​​‌ often considered through their‌ interleaved semantics, i.e. considering‌​‌ executions where, at each​​ step, a single transition​​​‌ fires. This is clearly‌ a miss, as Petri‌​‌ nets are a true​​ concurrency model. This paper​​​‌ revisits the semantics of‌ Petri nets as higher-dimensional‌​‌ automata (HDAs) as introduced​​ by van Glabbeek, which​​​‌ methodically take concurrency into‌ account. We extend the‌​‌ translation to include some​​ common features. We consider​​​‌ nets with inhibitor arcs,‌ under both concurrent semantics‌​‌ used in the literature,​​ and generalized self-modifying nets.​​​‌ Finally, we present a‌ tool that implements our‌​‌ translations.

Dynamic Partial Order​​ Reduction

Assessing the correctness​​​‌ of distributed and parallel‌ applications is notoriously difficult‌​‌ due to the complexity​​ of the concurrent behaviors​​​‌ and the difficulty to‌ reproduce bugs. In this‌​‌ context, Dynamic Partial Order​​ Reduction (DPOR) techniques have​​​‌ proved successful in exploiting‌ concurrency to verify applications‌​‌ without exploring all their​​ behaviors. However, they may​​​‌ lack of efficiency when‌ tracking non-systematic bugs of‌​‌ real size applications. In​​ this work 21,​​​‌ we suggest two adaptations‌ of the Optimal Dynamic‌​‌ Partial Order Reduction (ODPOR)​​ algorithm with a particular​​​‌ focus on bug finding‌ and explanation. The first‌​‌ adaptation is an out-of-order​​ version called RFS ODPOR​​​‌ which avoids being stuck‌ in uninteresting large parts‌​‌ of the state space.​​ Once a bug is​​​‌ found, the second adaptation‌ takes advantage of ODPOR‌​‌ principles to efficiently find​​ the origins of the​​​‌ bug.

Verification of blockchain‌ protocols

Blockchains use consensus‌​‌ protocols to reach agreement,​​ e.g., on the ordering​​​‌ of transactions. DAG-based consensus‌ protocols are increasingly adopted‌​‌ by blockchain companies to​​ reduce energy consumption and​​​‌ enhance security. These protocols‌ collaboratively construct a partial‌​‌ order of blocks (DAG​​ construction) and produce a​​​‌ linear sequence of blocks‌ (DAG ordering). Given the‌​‌ strategic significance of blockchains,​​ formal proofs of the​​​‌ correctness of key components‌ such as consensus protocols‌​‌ are essential. The contribution​​ 16 presents safety-verified specifications​​​‌ for five DAG-based consensus‌ protocols. Four of these‌​‌ protocols—DAG-Rider, Cordial Miners, Hashgraph,​​ and Eventual Synchronous BullShark—are​​​‌ well-established in the literature.‌ The fifth protocol is‌​‌ a minor variation of​​ Aleph, another well-established protocol.​​​‌ Our framework enables proof‌ reuse, reducing proof efforts‌​‌ by almost half. It​​​‌ achieves this by providing​ various independent, formally verified,​‌ specifications of DAG construction​​ and ordering variations, which​​​‌ can be combined to​ express all five protocols.​‌ We employ TLA+ for​​ specifying the protocols and​​​‌ writing their proofs, and​ the TLAPS proof system​‌ to automatically check the​​ proofs. Each TLA+ specification​​​‌ is relatively compact, and​ TLAPS efficiently verifies hundreds​‌ to thousands of obligations​​ within minutes. The significance​​​‌ of our work is​ two-fold: first, it supports​‌ the adoption of DAG-based​​ systems by providing robust​​​‌ safety assurances; second, it​ illustrates that DAG-based consensus​‌ protocols are amenable to​​ practical, reusable, and compositional​​​‌ formal methods.

8.3 New​ results on optimization of​‌ multi-agent systems

Participants: Nathalie​​ Bertrand, Aymeric Come​​​‌, Loïc Hélouët,​ Luca Paparazzo.

On​‌ this research axis, our​​ contributions this year concern​​​‌ the modelling and optimization​ of transportation networks, and​‌ the optimization of probabilistic​​ systems.

Modelling and optimizing​​​‌ transportation networks

With the​ application of optimization of​‌ urban transportation systems in​​ mind, we proposed several​​​‌ models for multi-agent systems​ and studied relevant verification​‌ questions.

In 23 we​​ consider a timed model​​​‌ tailored to study energy​ savings in transport networks.​‌ It focuses on regenerative​​ braking, a situation where​​​‌ the kinetic energy of​ a vehicle is transformed​‌ in electrical energy and​​ sent back in the​​​‌ electrical network. We first​ describe transport networks, and​‌ formalize their semantics as​​ timed runs of an​​​‌ equivalent network of timed​ automata (NTA). We then​‌ consider three problems. The​​ transfer existence problem checks​​​‌ whether a network has​ the ability to save​‌ energy. The ratio maximization​​ problem aims at computing​​​‌ the maximal ratio of​ energy saved by time​‌ unit in the long​​ run, and the threshold​​​‌ problem consists in verifying​ the existence of a​‌ strategy allowing the saving​​ of more energy than​​​‌ a fixed minima. We​ show that the transfer​‌ problem is a reachability​​ question in the region​​​‌ automaton of the NTA,​ which can be solved​‌ in PSPACE. The ratio​​ problem can be solved​​​‌ using Karp’s algorithm on​ a corner point abstraction​‌ for the NTA, yielding​​ an EXPTIME complexity. Finally,​​​‌ the threshold problem requires​ to address properties of​‌ elementary cycles and finite​​ paths of the corner-point​​​‌ automaton, yielding a PSPACE​ complexity.

In 17 we​‌ consider a new model​​ for multi-agent systems. This​​​‌ contribution introduces collaborative reachability​ games with energy constraints.​‌ In the considered arenas,​​ agents can spend or​​​‌ gain energy during moves,​ or share it with​‌ their peers if their​​ current position allows it.​​​‌ We study several variants​ of energy reachability games​‌ where agents move either​​ synchronously or asynchronously, and​​​‌ with/without constraints on energy​ transfers among peers. We​‌ show that these problems​​ have different complexities ranging​​​‌ from NP to EXPSPACE.​

Optimizating of probabilistic systems​‌

Markov chains and Markov​​ decision processes (MDPs) are​​​‌ well-established probabilistic models. We​ proposed two complementary contributions​‌ to the analysis of​​ MDPs: on the one​​​‌ hand an innovative depth-first​ search strategy, and on​‌ the other hand approximation​​ algorithms for large of​​ infinite-state models.

In the​​​‌ paper8 we consider‌ new algorithms to compute‌​‌ fixpoints in Markov Decison​​ Processes. Value and policy​​​‌ iteration are classical algorithms‌ to maximize the average‌​‌ discounted reward of an​​ MDP. They rely on​​​‌ a breadth-first exploration strategy‌ in the future of‌​‌ each state to update​​ its value and possibly​​​‌ change the action policy‌ at this state. This‌​‌ paper revisits this paradigm​​ and examines a depth-first​​​‌ search strategy. It reformulates‌ the average reward computation‌​‌ as an integral over​​ (future) paths that is​​​‌ better expressed in the‌ formalism of weighted automata.‌​‌ Policy evaluation can then​​ be solved by a​​​‌ Floyd-Warshall algorithm, which gathers‌ at once the rewards‌​‌ along possibly infinite runs.​​ This reformulation opens the​​​‌ way to new approximation‌ schemes for the value‌​‌ function. We show that​​ the same approach also​​​‌ gives access to other‌ quantities of interest, as‌​‌ the gradient of the​​ average reward with respect​​​‌ to model or policy‌ parameters, or the variance‌​‌ of the reward. The​​ behaviors and performances of​​​‌ this value estimation scheme‌ are illustrated on several‌​‌ benchmarks.

While finite Markov​​ models are well-understood, analyzing​​​‌ their infinite counterparts remains‌ a significant challenge. Decisiveness‌​‌ has proven to be​​ an elegant property for​​​‌ countable Markov chains: it‌ is general enough to‌​‌ be satisfied by several​​ natural classes of countable​​​‌ Markov chains, and it‌ is a sufficient condition‌​‌ for simple qualitative and​​ approximate quantitative model-checking algorithms​​​‌ to exist. In contrast,‌ existing works on the‌​‌ formal analysis of countable​​ MDPs usually rely on​​​‌ ad hoc techniques tailored‌ to specific classes. We‌​‌ provide in 25 a​​ general framework to analyze​​​‌ countable MDPs by extending‌ the notion of decisiveness.‌​‌ Compared to Markov chains,​​ MDPs exhibit extra non-determinism​​​‌ that can be resolved‌ in an adversarial or‌​‌ cooperative way, leading to​​ multiple natural notions of​​​‌ decisiveness. We show that‌ these notions enable the‌​‌ approximation of reachability and​​ safety probabilities in countable​​​‌ MDPs using simple model-checking‌ procedures. We then instantiate‌​‌ our generic approach to​​ two concrete classes of​​​‌ models inducing countable MDPs:‌ non-deterministic probabilistic lossy channel‌​‌ systems and partially observable​​ MDPs. This leads to​​​‌ an algorithm to approximately‌ compute safety probabilities in‌​‌ each of these classes.​​

9 Bilateral contracts and​​​‌ grants with industry

9.1‌ Bilateral contracts with industry‌​‌

Collaboration with Alstom Transport​​ - P22

Participants: Loïc​​​‌ Hélouët, Antoine Thébault‌.

DEVINE is involved‌​‌ in a long-term collaboration​​ with Alstom Transport on​​​‌ the topic of urbain‌ train systems regulation. Alstom‌​‌ and DEVINE jointly supervised​​ a CIFRE PhD on​​​‌ the topic of smart‌ traffic management (PhD of‌​‌ Antoine Thébault, ANRT grant​​ 2022-0444, 2022-2025). This PhD​​​‌ addressed the two following‌ objectives: optimize traffic management‌​‌ using concurrent models on​​ one hand, and learning​​​‌ techniques (neural networks training,‌ decision tree synthesis) on‌​‌ the other hand to​​ synthesize controllers.

10 Partnerships​​​‌ and cooperations

10.1 International‌ initiatives

10.1.1 Associate Teams‌​‌ in the framework of​​ an Inria International Lab​​​‌ or in the framework‌ of an Inria International‌​‌ Program

SINCRET

Participants: Loïc​​​‌ Germerie-Guizouarn, Thierry Jéron​, Sayan Mukerjee.​‌

  • Title:
    DST/Inria Associate Team​​ SINCRET, Scalable and INCREmental​​​‌ security monitoring and enforcement​ for Timed systems
  • Website:​‌
  • Partner​​ Institution:
    IIT Bhubaneswar, India​​​‌
  • Led by:
    Thierry Jéron​ (Inria) and Srinivas Pinisetty​‌ (IIT Bhubaneswar)
  • Objectives:
    The​​ objective of this project​​​‌ is to study the​ enforcement of timed properties​‌ for cyber physical systems​​ (CPS) in a synchronous​​​‌ context, and study their​ incremental composition. We first​‌ consider an enforcement monitor​​ synthesis framework for reactive​​​‌ CPS focused on discrete​ timed properties. This year​‌ we proposed a serial​​ composition scheme, delineate the​​​‌ sub-class of properties that​ are enforceable, and develop​‌ a serial composition scheme.​​ A prototype is developped​​​‌ by our partner and​ experimented on a case​‌ study of a swarm​​ of drones. A publication​​​‌ is under review for​ a journal. We also​‌ worked on Prompt Runtime​​ Enforcement in the untimed​​​‌ setting, and got a​ paper published at ATVA​‌ 2025 14. We​​ currently continue this work​​​‌ in the timed setting.​

10.1.2 Participation in other​‌ International Programs

IRL Relax​​

Participants: Nathalie Bertrand,​​​‌ Loïc Hélouët.

Several​ members of DEVINE have​‌ tight relationships with colleagues​​ at Chennai Mathematical Institute​​​‌ and IIT Bombay. For​ this reason, we are​‌ involved in the CNRS​​ IRL Relax, an​​​‌ Indo-French joint research unit​ dedicated to research in​‌ theoretical computer science and​​ in mathematics, their interactions​​​‌ and their applications.

10.2​ National initiatives

10.2.1 ANR​‌ projects

ANR MAVeriQ: Methods​​ of Analysis for Verification​​​‌ of Quantitative properties (2021-2025)​

Participants: Aymeric Côme,​‌ Loïc Hélouët, Nathalie​​ Bertrand.

  • Website:
  • Led by
    Aldric Degorre​ (IRIF); Local coordinator Éric​‌ Fabre.
  • Partners:
    IRIF, LMF,​​ Inria Rennes/IRISA, LACL, Verimag.​​​‌
  • Objectives:
    The objective of​ this project is to​‌ develop unified frameworks for​​ quantitative verification of timed,​​​‌ hybrid, and stochastic systems.​ We believe such a​‌ unification is possible because​​ common patterns are used​​​‌ in many cases. The​ project targets in particular:​‌ • systematization of quantitative​​ properties and their use​​​‌ cases • substantial progress​ in the algorithms of​‌ quantitative verification; • practical​​ methodology for stating and​​​‌ verifying quantitative properties of​ systems. The aim of​‌ MAVeriQ is to progress​​ towards this unification, by​​​‌ gathering skills on timed​ and stochastic systems and​‌ on quantitative verification under​​ a common roof, to​​​‌ jointly address open challenges​ in quantitative model-checking and​‌ quantitative validation. One such​​ challenge we will address​​​‌ is robustness of quantitative​ models, that is, resilience​‌ to small perturbations, which​​ is crucial for implementability.​​​‌ Unified methods developed in​ the project (such as​‌ robustness analysis and simulation​​ techniques) will be showcased​​​‌ in different case studies​ in the domain of​‌ CPS (in particular automotive​​ control), showing that such​​​‌ a system can be​ verified in different ways​‌ without leaving this framework.​​
ANR BisoUS: Better Synthesis​​​‌ for Underspecified Quantitative Systems​ (2023-2027)

Participants: Nathalie Bertrand​‌, Loïc Hélouët,​​ Nicolas Markey, Julie​​​‌ Parreaux, Ocan Sankur​.

  • Website:
  • Led​‌ by
    Didier Lime (LS2N);​​ Local coordinator Nathalie Bertrand​​ .
  • Partners:
    LS2N, Inria​​​‌ Rennes/IRISA, LIPN, LMF.
  • Objectives:‌
    When designing complex and‌​‌ critical systems (planes, autonomous​​ vehicles, etc.), it is​​​‌ crucial to be able‌ to give guarantees that‌​‌ the system works as​​ intended, which is often​​​‌ done through comprehensive testing.‌ The goal of project‌​‌ BisoUS is to provide​​ stronger guarantees, based on​​​‌ formal methods, and to‌ detect problems as early‌​‌ as possible: solving them​​ is then easier and​​​‌ cheaper. Unfortunately this is‌ a hard problem because‌​‌ some design choices may​​ not have been done​​​‌ yet, and some key‌ features (e.g. speed of‌​‌ a CPU) are then​​ not known precisely enough.​​​‌ In project BisoUS we‌ develop formal methods, based‌​‌ on model-checking and synthesis​​ to work with expressive​​​‌ modelling formalisms encompassing parameters,‌ cost/rewards, and games on‌​‌ graphs to meet those​​ challenges.
ANR PaVeDyS: Parametric​​​‌ Verification of Dynamic Distributed‌ Systems (2024-2027)

Participants: Nathalie‌​‌ Bertrand, Loïc Germerie-Guizouarn​​, Thierry Jéron,​​​‌ Ocan Sankur.

  • Website:‌
  • Led by
    Radu‌​‌ Iosif (Verimag); Local coordinator​​ Nathalie Bertrand .
  • Partners:​​​‌
    Verimag, Inria Rennes, IRIF,‌ LaBRI.
  • Objectives:
    Applications of‌​‌ distributed systems are omnipresent.​​ They allow sharing resources​​​‌ and data. They are‌ used to coordinate activities‌​‌ across multiple nodes, as​​ in geographically distributed systems.​​​‌ Furthermore, they increase the‌ resilience of systems through‌​‌ fault tolerance, availability, and​​ recovery mechanisms. Designing, understanding,​​​‌ and validating distributed systems‌ are challenging because of‌​‌ the huge number of​​ interactions between components, some​​​‌ potentially leading to unpredictable‌ scenarios. Early detection of‌​‌ design errors is not​​ only crucial for financial​​​‌ reasons, but it is‌ often the only feasible‌​‌ way to find critical​​ errors. The methods for​​​‌ ensuring the correctness of‌ distributed systems are not‌​‌ yet mature. This is​​ particularly the case for​​​‌ the mechanized reasoning methods‌ that we propose to‌​‌ develop in this project.​​
Défi FRAIME (2025-2029)

Participants:​​​‌ Nathalie Bertrand, Ocan‌ Sankur.

  • Led by‌​‌
    Nathalie Bertrand (DEVINE) and​​ David Mentré (MERCE).
  • Partners:​​​‌
    Gallinette, EPICURE, ARGO, DiverSE,‌ DEVINE, MERCE.
  • Objectives:
    Together‌​‌ with MERCE (Mitsubishi Electric​​ Research Center Europe), Inria​​​‌ launched mid september 2025‌ a joint Défi. FRAIME‌​‌ gathers five Inria teams​​ (DEVINE, Epicure, Gallinette, DiverSE,​​​‌ ARGO) and two research‌ teams from MERCE Digital‌​‌ Information Systems division (Information​​ and Network Systems and​​​‌ Synergistic Autonomous Systems). In‌ FRAIME, we propose to‌​‌ explore on the one​​ hand how Formal Methods​​​‌ can provide guarantees on‌ AI systems, and on‌​‌ the other hand how​​ AI can help Formal​​​‌ Methods to be more‌ efficient and easier to‌​‌ use by practitioners. The​​ vision is to intertwine​​​‌ Formal Methods and AI‌ to efficiently design safe‌​‌ systems. The project is​​ driven by industrial use​​​‌ cases in the expertise‌ of MERCE and targets‌​‌ the following four challenging​​ objectives: user-assisted factory automation​​​‌ code generation, correctness of‌ C programs, safe and‌​‌ optimal system control, easier​​ and more automated proof​​​‌ assistant.

10.2.2 National Informal‌ Collaborations

The team collaborates‌​‌ with the following researchers:​​

  • Patricia Bouyer (LMF, ENS​​​‌ Paris-Saclay) on quantitative aspects‌ of verification and game‌​‌ models for parameterized systems;​​​‌
  • Luc Dartois (FEMTO-ST, Université​ de Besançon) and Paul​‌ Gastin (LMF, ENS Paris-Saclay)​​ on transducer models and​​​‌ verification of transformations;

11​ Dissemination

11.1 Promoting scientific​‌ activities

11.1.1 Scientific events:​​ selection

Chair of conference​​​‌ program committees
  • Nathalie Bertrand​ serves as PC chair​‌ for the international conference​​ FoSSaCS'26. This activity​​​‌ spans from summer 2025​ (starting with identifying PC​‌ members) to spring 2026​​ (ending with the conference​​​‌ being held).
Member of​ the conference program committees​‌
  • Nathalie Bertrand has served​​ on the PC of​​​‌ the international conferences FoSSaCS'25​ and MFCS'25.
  • Loïc Hélouët​‌ has served as PC​​ member of the international​​​‌ conferences Petri Nets 2025,​ NETYS 2025, and ATMOS​‌ 2025.
  • Thierry Jéron was​​ in the program committee​​​‌ of SAC-SVT'25.
  • Julie Parreaux​ was in the program​‌ comittee of ICFEM'25.
Reviewer​​

 

DEVINE members regularly write​​​‌ reviews for the central​ international conferences of our​‌ field: LICS, ICALP, CAV,​​ Concur, FSTTCS, etc.

11.1.2​​​‌ Journal

Member of the​ editorial boards

 

Nathalie Bertrand​‌ is an editorial board​​ member for Journal of​​​‌ Logical and Algebraic Methods​ in Programming (JLAMP) and​‌ for Theoretical Computer Science​​ (TCS).

Reviewer reviewing activities​​​‌

 

All members of the​ team reviewed a number​‌ of papers for international​​ journals in their field​​​‌ of expertise.

11.1.3 Leadership​ within the scientific community​‌

Nathalie Bertrand is the​​ co-head of the French​​​‌ working group on verification​ of GDR-IFM: GT Vérif​‌.

11.1.4 Research administration​​

  • Nathalie Bertrand is member​​​‌ of the Formation Spécialisée​ de Site at Inria​‌ center of the University​​ of Rennes.
  • Nathalie Bertrand​​​‌ is co-head of IRISA​ and Inria Rennes gender​‌ equality committee, and member​​ of Inria's gender equality​​​‌ and equal opportunities committee.​
  • Loïc Hélouët is the​‌ president of the Commission​​ Personnel (Temporary staff committee)​​​‌ at Inria Rennes, responsible​ for evaluation of hirings​‌ on non-permanent positions.
  • Loïc​​ Hélouët is elected member​​​‌ of the "Formation Specialisée​ (Health and security committee)"​‌ of Inria and secretary​​ of this committee.
  • Thierry​​​‌ Jéron is référent chercheur​ for Inria Rennes and​‌ IRISA.

11.2 Teaching Supervision​​ Juries Educational and pedagogical​​​‌ outreach

11.2.1 Teaching

Almost​ all members of DEVINE​‌ do teach.

  • Licence: Nathalie​​ Bertrand , Algorithms II,​​​‌ 18h, ENS Rennes;

  • BUT:​ Loïc Germerie Guizouarn, network​‌ and cyber-security classes, 246h,​​ IUT Saint-Malo;

  • Licence: Aline​​​‌ Goeminne, Algorithms I, 19.5h,​ ENS Rennes;
  • Master: Aline​‌ Goeminne, Algorithms, 9h, Agrégation,​​ ENS Rennes;
  • Master: Aline​​​‌ Goeminne, Computability and Complexity,​ 10.5h, Agrégation, ENS Rennes;​‌
  • Master: Aline Goeminne, Formal​​ languages, 12h, Agrégation, ENS​​​‌ Rennes;
  • Master: Aline Goeminne,​ Algorithms and data structures,​‌ 16h, Agrégation, ENS Rennes.​​

  • Master: Loïc Hélouët ,​​​‌ Algorithms and proofs, 16h,​ Agrégation, ENS Rennes;
  • Licence:​‌ Loïc Hélouët , Algorithms​​ and Java, 40h, INSA​​​‌ Rennes;

  • Licence: Julie Parreaux​ , Algorithms 2, 21h,​‌ ENS Rennes;
  • Licence: Julie​​ Parreaux , Formals Tools​​​‌ for Computer scientists, 52h,​ ISTIC, Université de Rennes;​‌
  • Licence: Julie Parreaux ,​​ Algorithms and Complexity, 47h,​​​‌ ISTIC, Université de Rennes;​
  • Licence: Julie Parreaux ,​‌ Programming 2, 24h, ISTIC,​​ Université de Rennes;
  • Licence:​​​‌ Julie Parreaux , Logic,​ 16.5h, ISTIC, Université de​‌ Rennes;
  • Master: Julie Parreaux​​ , Validation and Verification,​​ 16.5h, ISTIC, Université de​​​‌ Rennes.

11.2.2 Supervision

PhD‌ Students
  • PhD in Progress‌​‌
    • Mathieu Laurent , on​​ Efficient verification of asynchronous​​​‌ distributed systems, started in‌ October 2023, supervised by‌​‌ Martin Quinson (Magellan) and​​ Thierry Jéron ;
    • Luc​​​‌ Lapointe (ENS Paris-Saclay), AGPR‌ ENS Paris-Saclay, on concurrent‌​‌ games with parameterized number​​ of participants, started in​​​‌ September 2023, supervised by‌ Nathalie Bertrand and Patricia‌​‌ Bouyer (LMF);
    • Luca Paparazzo​​ , ENS Grant, on​​​‌ Quantitative games in timed‌ systems, application to energy‌​‌ savings in urban transport,​​ started in October 2024,​​​‌ supervised by Nathalie Bertrand‌ and Loïc Hélouët ;‌​‌
    • Aymeric Come , on​​ Approximation methods for the​​​‌ soundness of control laws‌ derived by machine learning,‌​‌ started in December 2022,​​ supervised by Eric Fabre​​​‌ and Loïc Hélouët ;‌
    • Pranav Ghorpade (Univ. Sydney),‌​‌ on Verification of distributed​​ algorithms within blockchains, started​​​‌ in October 2024, supervised‌ by Nathalie Bertrand and‌​‌ Sasha Rubin (Univ. Sydney).​​
    • Victorien Desbois , on​​​‌ Heuristic search algorithms for‌ vehicle rescheduling problem, started‌​‌ in December 2023, supervised​​ by Ocan Sankur, François​​​‌ Schwarzentruber, Cédric Pelois (NewLogUp).‌
  • Past PhDs
    • Antoine Thébault‌​‌ , CIFRE Grant, on​​ Efficient learning techniques for​​​‌ traffic management in Transport‌ Networks, started in September‌​‌ 2022, supervised by Kenza​​ Saiah (Alstom Transport) and​​​‌ Loïc Hélouët . PhD‌ not defended.
Master Students‌​‌
  • Maëlle Gautrin , M2​​ student at ENS Paris-Saclay,​​​‌ has been supervised for‌ her M2 research internship‌​‌ on the Robustness in​​ Weighted Timed Games by​​​‌ Nathalie Bertrand and Julie‌ Parreaux .
Undergraduate Students‌​‌
  • Nowar Kazem , L3​​ student at ENS Rennes,​​​‌ has been supervised by‌ Loïc Germerie-Guizouarn and Thierry‌​‌ Jéron on the Decidability​​ of deadlock-freedom of RSC​​​‌ systems.

11.2.3 Juries

PhD‌ committees
  • Nathalie Bertrand was‌​‌ on the PhD defense​​ jury of Alexandre Terefenko​​​‌ (UMONS, Belgium, march 2025),‌ James Main (UMONS, Belgium,‌​‌ sept. 2025), Lucie Guillou​​ (Université Paris Cité, sept.​​​‌ 2025). She was reviewer‌ and took part to‌​‌ the jury of the​​ PhD of Mathias Déhais​​​‌ (Université Caen Normandie, nov.‌ 2025).
  • Loïc Hélouët was‌​‌ reviewer in the PhD​​ jury of Sarah Larroze​​​‌ Jardiné at LABRI in‌ Dec. 2025.
Hiring committees‌​‌
  • In 2025, Nathalie Bertrand​​ has been on the​​​‌ hiring committee of a‌ Professor position at INSA‌​‌ Rennes, a Professor position​​ at ENS Paris-Saclay, and​​​‌ on the admissibility jury‌ of CRCN-ISFP competition at‌​‌ Inria centre at the​​ University Grenoble Alpes.

11.3​​​‌ Popularization

11.3.1 Participation in‌ Live events

  • Loïc Hélouët‌​‌ contributed to the organization​​ of a seminar on​​​‌ Livestorm for PhD and‌ Postdocs at Inria Rennes‌​‌ (1h each, 40-50 participants),​​ dedicated to Work environment​​​‌ for PhD students.

11.3.2‌ Others science outreach relevant‌​‌ activities

  • Loïc Hélouët contributed​​ to the CHICHE! program​​​‌ (15 classes visited).

12‌ Scientific production

12.1 Major‌​‌ publications

12.2 Publications of the​‌ year

International journals

International peer-reviewed​​​‌ conferences

Conferences without proceedings

Scientific book chapters​

Edition (books, proceedings, special​​​‌ issue of a journal)​

  • 26 proceedingsPrinciples of​‌ Formal Quantitative Analysis: Essays​​ Dedicated to Christel Baier​​​‌ on the Occasion of​ Her 60th Birthday.​‌Principles of Formal Quantitative​​ Analysis15760Lecture Notes​​​‌ in Computer ScienceAarhus​ (Danemark), DenmarkSpringer Nature​‌ Switzerland2025HALDOI​​

12.3 Cited publications

  • 27​​​‌ articleR.Rajeev Alur​ and D. L.David​‌ L. Dill. A​​ Theory of Timed Automata​​​‌.Theoretical Computer Science​12621994,​‌ 183-235DOIback to​​ text
  • 28 articleK.​​​‌Krzysztof Apt and D.​ C.Dexter C. Kozen​‌. Limits for automatic​​ verification of finite-state concurrent​​​‌ systems.Information Processing​ Letters226May​‌ 1986, 307-309DOI​​back to text
  • 29​​​‌ inproceedingsS.Stavros Aronis​, B.Bengt Jonsson​‌, M.Magnus L\aang​​ and K.Konstantinos Sagonas​​​‌. Optimal Dynamic Partial​ Order Reduction with Observers​‌.TACAS'1810806LNCS​​Springer2018, 229-248​​​‌DOIback to text​
  • 30 inbookE.Ezio​‌ Bartocci, Y.Yliès​​ Falcone, A.Adrian​​​‌ Francalanza and G.Giles​ Reger. Introduction to​‌ Runtime Verification.Lectures​​ on Runtime Verification: Introductory​​​‌ and Advanced TopicsSpringer​2018, 1--33DOI​‌back to text
  • 31​​ inproceedingsG.Gerd Behrmann​​​‌, A.Alexandre David​, K. G.Kim​‌ Guldstrand Larsen, J.​​John H\aa}kansson, P.​​Paul Pettersson, W.​​​‌Wang Yi and M.‌Martijn Hendriks. UPPAAL~4.0‌​‌.QEST'06IEEE Comp.​​ Soc. PressSeptember 2006​​​‌, 125-126DOIback‌ to text
  • 32 inproceedings‌​‌N.Nathalie Bertrand,​​ P.Paulin Fournier and​​​‌ A.Arnaud Sangnier.‌ Playing with Probabilities in‌​‌ Reconfigurable Broadcast Networks.​​FoSSaCS'148412LNCSSpringer​​​‌2014, 134--148DOI‌back to text
  • 33‌​‌ inproceedingsN.Nathalie Bertrand​​, I.Igor Konnov​​​‌, M.Marijana Lazic‌ and J.Josef Widder‌​‌. Verification of Randomized​​ Consensus Algorithms Under Round-Rigid​​​‌ Adversaries.CONCUR'19140‌LIPIcsLeibniz-Zentrum für Informatik‌​‌2019, 33:1--33:15DOI​​back to text
  • 34​​​‌ inproceedingsN.Nathalie Bertrand‌, M.Marijana Lazic‌​‌ and J.Josef Widder​​. A~Reduction Theorem for​​​‌ Randomized Distributed Algorithms Under‌ Weak Adversaries.VMCAI'21‌​‌12597LNCSSpringer2021​​, 219--239DOIback​​​‌ to text
  • 35 inproceedings‌N.Nathalie Bertrand,‌​‌ B.Bastien Thomas and​​ J.Josef Widder.​​​‌ Guard Automata for the‌ Verification of Safety and‌​‌ Liveness of Distributed Algorithms​​.CONCUR'21203LIPIcs​​​‌Leibniz-Zentrum für Informatik2021‌, 15:1--15:17DOIback‌​‌ to text
  • 36 inproceedings​​B.Borzoo Bonakdarpour,​​​‌ P.Pavithra Prabhakar and‌ C.César Sánchez.‌​‌ Model Checking Timed Hyperproperties​​ in Discrete-Time Systems.​​​‌NFM'2012229LNCSSpringer‌2020, 311--328DOI‌​‌back to text
  • 37​​ articleP.Patricia Bouyer​​​‌, L.Léo Henry‌, S.Samy Jaziri‌​‌, T.Thierry Jéron​​ and N.Nicolas Markey​​​‌. Diagnosing timed automata‌ using timed markings.‌​‌International Journal on Software​​ Tools for Technology Transfer​​​‌232April 2021‌, 229-253DOIback‌​‌ to text
  • 38 article​​P.Patricia Bouyer,​​​‌ O.Orna Kupferman,‌ N.Nicolas Markey,‌​‌ B.Bastien Maubert,​​ A.Aniello Murano and​​​‌ G.Giuseppe Perelli.‌ Reasoning about Quality and‌​‌ Fuzziness of Strategic Behaviours​​.ACM Transactions on​​​‌ Computational LogicTo~appear2023‌back to text
  • 39‌​‌ inproceedingsR.Romain Brenguier​​, G. A.Guillermo​​​‌ A. Pérez, J.-F.‌Jean-François Raskin and O.‌​‌Ocan Sankur. Admissibility​​ in Quantitative Graph Games​​​‌.FSTTCS'1665LIPIcs‌Leibniz-Zentrum für Informatik2016‌​‌, 42:1--42:14DOIback​​ to text
  • 40 article​​​‌K.Krishnendu Chatterjee,‌ T. A.Thomas A.‌​‌ Henzinger and N.Nir​​ Piterman. Strategy Logic​​​‌.Information and Computation‌2086June 2010‌​‌, 677-693DOIback​​ to text
  • 41 inproceedings​​​‌E.E Clarke,‌ D.D Long and‌​‌ K.K McMillan.​​ Compositional model checking.​​​‌LICS'89IEEE Comp. Soc.‌ Press1989, 353--362‌​‌back to text
  • 42​​ inproceedingsM. R.Michael​​​‌ R. Clarkson, B.‌Bernd Finkbeiner, M.‌​‌Masoud Koleini, K.​​ K.Kristopher K. Micinski​​​‌, M. N.Markus‌ N. Rabe and C.‌​‌César Sánchez. Temporal​​ Logics for Hyperproperties.​​​‌POST'148414LNCSSpringer‌2014, 265--284DOI‌​‌back to text
  • 43​​ articleM. R.Michael​​​‌ R. Clarkson and F.‌ B.Fred B. Schneider‌​‌. Hyperproperties.Journal​​ of Computer Security18​​​‌62010, 1157--1210‌DOIback to text‌​‌
  • 44 inproceedingsE.Emily​​​‌ Clement, T.Thierry​ Jéron, N.Nicolas​‌ Markey and D.David​​ Mentré. Computing maximally-permissive​​​‌ strategies in acyclic timed​ automata.FORMATS'2012288​‌LNCSSpringerSeptember 2020​​, 111-126DOIback​​​‌ to text
  • 45 inproceedings​K. E.Katherine E.​‌ Coons, M.Madan​​ Musuvathi and K. S.​​​‌Kathryn S. McKinley.​ Bounded partial-order reduction.​‌OOPSLA'13ACM~Press2013,​​ 833--848DOIback to​​​‌ text
  • 46 articleA.​Andrea D'Ariano, M.​‌Marco Pranzo and I.​​ A.Ingo A. Hansen​​​‌. Conflict Resolution and​ Train Speed Coordination for​‌ Solving Real-Time Timetable Perturbations​​.IEEE Transactions on​​​‌ Intelligent Transportation Systems8​22007, 208-222​‌back to text
  • 47​​ inproceedingsL.Luc Dartois​​​‌, P.Paulin Fournier​, I.Ismaël Jecker​‌ and N.Nathan Lhote​​. On Reversible Transducers​​​‌.44th International Colloquium​ on Automata, Languages, and​‌ Programming, ICALP 2017, Warsaw,​​ Poland, July 10-14, 2017​​​‌80LIPIcsSchloss Dagstuhl​ - Leibniz-Zentrum für Informatik​‌2017, 113:1--113:12URL:​​ https://doi.org/10.4230/LIPIcs.ICALP.2017.113DOIback to​​​‌ text
  • 48 inproceedingsC.​Christian Dehnert, D.​‌Daniel Gebler, M.​​Michele Volpato and D.​​​‌ N.David N. Jansen​. On Abstraction of​‌ Probabilistic Systems.ROCKS'12​​8453LNCSSpringer2012​​​‌, 87--116DOIback​ to text
  • 49 article​‌A.Andrea D’Ariano,​​ D.Dario Pacciarelli and​​​‌ M.Marco Pranzo.​ A branch-and-bound algorithm for​‌ scheduling trains in a​​ railway network.European​​​‌ Journal of Operational Research​18322007,​‌ 643–657back to text​​
  • 50 inproceedingsS.Stefan​​​‌ Edelkamp, V.Viktor​ Schuppan, D.Dragan​‌ Bosnacki, A.Anton​​ Wijs, A.Ansgar​​​‌ Fehnker and H.Husain​ Aljazzar. Survey on​‌ Directed Model Checking.​​MoChArt'085348LNCSSpringer​​​‌2008, 65--89DOI​back to text
  • 51​‌ inproceedingsA. B.Andreas​​ Berre Eriksen, C.​​​‌Chao Huang, J.​Jan Kildebogaard, H.​‌Harry Lahrmann, K.​​ m.Ki mG. Larsen​​​‌, M.Marco Muñiz​ and J. H.Jakob​‌ Haahr Taankvist. Uppaal​​ Stratego for Intelligent Traffic​​​‌ Lights.12th ITS​ European Congress2017back​‌ to text
  • 52 inproceedings​​C.Cormac Flanagan and​​​‌ P.Patrice Godefroid.​ Dynamic partial-order reduction for​‌ model checking software.​​POPL'05ACM~Press2005,​​​‌ 110--121DOIback to​ text
  • 53 inproceedingsS.​‌Susanne Graf and H.​​Hassen Saidi. Construction​​​‌ of abstract state graphs​ with~PVS.CAV'971254​‌LNCSSpringer1997,​​ 72--83back to text​​​‌
  • 54 inproceedingsK.Klaus​ Havelund, A.Arne​‌ Skou, K. G.​​Kim Guldstrand Larsen and​​​‌ K.Kristian Lund.​ Formal modeling and analysis​‌ of an audio/video protocol:​​ An industrial case study​​​‌ using UPPAAL.Proceedings​ Real-Time Systems SymposiumIEEE​‌1997, 2--13back​​ to text
  • 55 article​​​‌L.Loïc Hélouët,​ N.Nicolas Markey and​‌ R.Ritam Raha.​​ Reachability games with relaxed​​​‌ energy constraints.Information​ and Computation285~(Part~B)May​‌ 2022DOIback to​​ text
  • 56 inproceedingsL.​​​‌Léo Henry, T.​Thierry Jéron and N.​‌Nicolas Markey. Active​​ Learning of Timed Automata​​ with Unobservable Resets.​​​‌FORMATS'2012288LNCSSpringer‌September 2020, 144-160‌​‌DOIback to text​​
  • 57 articleL.Léo​​​‌ Henry, T.Thierry‌ Jéron and N.Nicolas‌​‌ Markey. Control strategies​​ for off-line testing of​​​‌ timed systems.Formal‌ Methods in System Design‌​‌2023DOIback to​​ textback to text​​​‌
  • 58 miscF.Frédéric‌ Herbreteau and G.Gerald‌​‌ Point. TChecker.​​https://github.com/ticktac-project/tchecker2019back to​​​‌ textback to text‌
  • 59 inproceedingsH.Holger‌​‌ Hermanns, B.Björn​​ Wachter and L.Lijun​​​‌ Zhang. Probabilistic CEGAR‌.CAV'085123LNCS‌​‌Springer2008, 162--175​​DOIback to text​​​‌
  • 60 articleA.András‌ Horváth, M.Marco‌​‌ Paolieri, L.Lorenzo​​ Ridi and E.Enrico​​​‌ Vicario. Transient analysis‌ of non-Markovian models using‌​‌ stochastic state classes.​​Performance Evaluation697-8​​​‌2012, 315--335back‌ to text
  • 61 inproceedings‌​‌T.Thierry Jéron,​​ N.Nicolas Markey,​​​‌ D.David Mentré,‌ R.Reiya Noguchi and‌​‌ O.Ocan Sankur.​​ Incremental methods for checking​​​‌ real-time consistency.FORMATS'20‌12288LNCSSpringerSeptember‌​‌ 2020, 249-264DOI​​back to text
  • 62​​​‌ inproceedingsK.Kai Lampka‌, S.Simon Perathoner‌​‌ and L.Lothar Thiele​​. Analytic real-time analysis​​​‌ and timed automata: a‌ hybrid method for analyzing‌​‌ embedded real-time systems.​​Proceedings of the seventh​​​‌ ACM international conference on‌ Embedded software2009,‌​‌ 107--116back to text​​
  • 63 inproceedingsD. J.​​​‌Daniel J. Lehmann and‌ M. O.Michael O.‌​‌ Rabin. On the​​ Advantages of Free Choice:​​​‌ A~Symmetric and Fully Distributed‌ Solution to the Dining‌​‌ Philosophers Problem.POPL'81​​ACM~Press1981, 133--138​​​‌DOIback to text‌
  • 64 inproceedingsZ.Ziqing‌​‌ Luo, M.Manchun​​ Zheng and S. F.​​​‌Stephen F. Siegel.‌ Verification of MPI programs‌​‌ using CIVL.EuroMPI'17​​ACM~Press2017, 6:1--6:11​​​‌DOIback to text‌
  • 65 miscF.Fabio‌​‌ Martinelli, F.Francesco​​ Mercaldo, A.Antonella​​​‌ Santone, C.Christina‌ Tavolato-Wötzl and P.Paul‌​‌ Tavolato. Timed Automata​​ Networks for SCADA Attacks​​​‌ Real-Time Mitigation.2019‌back to text
  • 66‌​‌ inproceedingsT. A.The​​ Anh Pham, T.​​​‌Thierry Jéron and M.‌Martin Quinson. Unfolding-Based‌​‌ Dynamic Partial Order Reduction​​ of Asynchronous Distributed Programs​​​‌.FORTE 201911535‌LNCSSpringer2019,‌​‌ 224--241DOIback to​​ text
  • 67 bookM.​​​‌ L.Martin L. Puterman‌. Markov decision processes:‌​‌ discrete stochastic dynamic programming​​.John Wiley~&~Sons2014​​​‌back to text
  • 68‌ inproceedingsC.César Rodríguez‌​‌, M.Marcelo Sousa​​, S.Subodh Sharma​​​‌ and D.Daniel Kroening‌. Unfolding-based Partial Order‌​‌ Reduction.CONCUR'1542​​LIPIcsLeibniz-Zentrum für Informatik​​​‌2015, 456--469DOI‌back to text
  • 69‌​‌ inproceedingsV.Victor Roussanaly​​, O.Ocan Sankur​​​‌ and N.Nicolas Markey‌. Abstraction Refinement Algorithms‌​‌ for Timed Automata.​​CAV'1911561LNCSSpringer​​​‌July 2019, 22-40‌DOIback to text‌​‌
  • 70 inproceedingsO.Ocan​​ Sankur and B.Bastien​​​‌ Thomas. PyLTA: A‌ Verification Tool for Parameterized‌​‌ Distributed Algorithms.TACAS'23​​​‌13994LNCSSpringer2023​back to text
  • 71​‌ articleH. A.Herbert​​ A. Simon. Rational​​​‌ Choice and the Structure​ of the Environment.​‌Psychological Review632​​1956, 129-138back​​​‌ to text
  • 72 article​A.Akihiro Tamatsukuri and​‌ T.Tatsuji Takahashi.​​ Guaranteed satisficing and finite​​​‌ regret: Analysis of a​ cognitive satisficing value function​‌.Biosystems1802019​​, 46-53back to​​​‌ text
  • 73 articleM.​Mark Utting, A.​‌Alexander Pretschner and B.​​Bruno Legeard. A~taxonomy​​​‌ of model-based testing approaches​.Software Testing, Verification​‌ and Reliability225​​2012, 297--312DOI​​​‌back to text
  1. 1​We almost always target​‌ high-level models for distributed​​ systems rather than the​​​‌ systems themselves. Bug finding​ in MPI (Message passing​‌ interface) programs —see Section​​ 3.2— is an​​​‌ exception.