CV
Contact Information
| Name | Marta Grobelna |
| Professional Title | Doctoral Researcher, Algorithm Development and Optimization |
| marta.grobelna@tum.de | |
| Location | Munich, Germany |
| Website | https://www.marta-grobelna.com |
Profile
Doctoral researcher at the Technical University of Munich developing algorithms for concurrent stochastic games and Markov decision processes with formal precision guarantees. I independently developed, proved correct, and implemented a novel value-iteration algorithm in the Java-based PRISM model checker, using linear and quadratic programming (AMPL, Gurobi) as core tools. At Fraunhofer IKS, I translated industry customers’ goals (BMW, Huawei) into mathematical models and working prototypes, and I maintain and migrate a production Python web platform (Automata Tutor). Next, I want to bring this experience to industry and tackle challenging real-world problems.
Experience
-
2022 - present Munich, Germany
Doctoral Researcher, Algorithm Development
Technical University of Munich (TUM)
- Design algorithms for stochastic games and Markov decision processes (MDPs), combining mathematical analysis, correctness proofs, software implementation, and computational evaluation.
- Independently designed, proved correct, and implemented in Java the first value-iteration algorithm for concurrent stochastic games with guaranteed precision for maximum-reachability objectives, and integrated it into the PRISM probabilistic model checker (LICS 2025).
- Formulated optimization problems arising from concurrent stochastic games in AMPL and solved them using Gurobi and other optimizers through the AMPL Java API.
- Co-authored a paper on hot-starting and incremental value iteration for MDPs that reuses previous results when a model changes, with formal precision guarantees (under review).
- Currently developing bounded strategy iteration and a new quadratic-programming formulation for concurrent stochastic games, and extending the methods to partially observable MDPs (POMDPs).
-
2019 - 2022 Munich, Germany
Research Associate, Applied R&D for Autonomous Systems
Fraunhofer Institute for Cognitive Systems (IKS)
- BMW project: worked directly with the customer to turn the question “is this overtaking maneuver safe?” into a precise mathematical problem (reachable-set computation from vehicle dynamics via flow-pipe construction, checking safety constraints specified in temporal logic for hybrid automata). Delivered a software prototype and was first author of the resulting EDCC 2021 paper.
- Huawei project: analyzed safety properties of sensor-fusion algorithms for autonomous driving and contributed to the white paper Trustworthy AI for Intelligent Traffic Systems.
- Coordinated work packages, clarified customer requirements, and presented results to industry stakeholders.
-
2018 - 2019 Aachen, Germany
Student Assistant, Software and Verification Tools
RWTH Aachen University
- Benchmarked verification tools and contributed C++ implementation and project documentation.
Technical Skills
Projects
-
Automata Tutor
Web-based platform for automated assessment of exercises in automata theory (Python/Flask, MySQL, pytest, CI/CD).
- Re-implemented the entire frontend in Python/Flask and migrated the database from H2 to MySQL, including a redesign of the SQL schema.
- Built the automated test infrastructure (pytest fixtures, mock databases) within a collaborative Git and CI/CD workflow.
- Currently guiding students through the migration of the backend.
Publications
-
2025 Stopping Criteria for Value Iteration on Concurrent Stochastic Reachability and Safety Games
Marta Grobelna, Jan Křetínský, Maximilian Weininger
LICS 2025
-
2021 Dynamic Risk Management for Safely Automating Connected Driving Maneuvers
Marta Grobelna, Joao-Vitor Zacchi, Philipp Schleiss, Simon Burton
EDCC 2021
-
2021 Trustworthy AI for Intelligent Traffic Systems (ITS)
Stefano Bortoli, Margherita Grossi, Simon Burton, Marta Grobelna, Yuki Hagawara, Philipp Schleiß
Fraunhofer IKS white paper
Education
-
2022 - present Munich, Germany
PhD
Technical University of Munich
Computer Science
- Expected completion in February 2027.
-
2017 - 2019 Aachen, Germany
M.Sc.
RWTH Aachen University
Computer Science
- Master’s thesis: Integration of hybrid systems verification methods in Matlab
-
2013 - 2017 Aachen, Germany
Leadership and Teaching
- Coordinated approximately 15 teaching assistants for a university course with 750+ students.
- Supervised four Bachelor’s and two Master’s theses.
- Regularly present my research at international conferences.