Timo P. Gros
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Timo P. Gros has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Verena Wolf
Saarland University
6 shared publications - Jörg Hoffmann
Saarland University Saarbrücken Germany
5 shared publications - Michaela Klauck
Robert Bosch GmbH
5 shared publications Daniel Höller
Saarland University, Saarland Informatics Campus
3 shared publications- Holger Hermanns
Saarland University
3 shared publications Jörg Hoffmann
Saarland University, Saarland Informatics Campus
3 shared publications
Latest Publications
- 2 October 2025
- Published by Springer Nature
- p. 134-156
PyDSMC: Statistical Model Checking for Neural Agents Using the Gymnasium Interface
- 2 October 2025
- Published by Springer Nature
- p. 134-156
PyDSMC: Statistical Model Checking for Neural Agents Using the Gymnasium Interface
- 18 November 2024
- Published by Springer Nature
- p. 18-38
Safe Reinforcement Learning Through Regret and State Restorations in Evaluation Stages
- 18 November 2024
- Published by Springer Nature
- p. 18-38
Safe Reinforcement Learning Through Regret and State Restorations in Evaluation Stages
- 24 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (11) , 17852-17864
Motion Primitives as the Action Space of Deep Q-Learning for Planning in Autonomous Driving
Motion planning for autonomous vehicles is commonly implemented via graph-search methods, which pose limitations to the model accuracy and environmental complexity that can be handled under real-time constraints. In contrast, reinforcement learning, specifically the deep Q-learning (DQL) algorithm, provides an interesting alternative for real-time solutions. Some approaches, such as the (), model the RL-action space by quantizing the continuous control inputs. Here, we propose to use motion primitives, which encode continuous-time nonlinear system behavior as the action space. The novel methodology of motion primitives-DQL planning is evaluated in a numerical example using a single-track vehicle model and different planning scenarios. We show that our approach outperforms a state-of-the-art graph-search method in computation time and probability of reaching the goal.
- 24 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (11) , 17852-17864
Motion Primitives as the Action Space of Deep Q-Learning for Planning in Autonomous Driving
Motion planning for autonomous vehicles is commonly implemented via graph-search methods, which pose limitations to the model accuracy and environmental complexity that can be handled under real-time constraints. In contrast, reinforcement learning, specifically the deep Q-learning (DQL) algorithm, provides an interesting alternative for real-time solutions. Some approaches, such as the (), model the RL-action space by quantizing the continuous control inputs. Here, we propose to use motion primitives, which encode continuous-time nonlinear system behavior as the action space. The novel methodology of motion primitives-DQL planning is evaluated in a numerical example using a single-track vehicle model and different planning scenarios. We show that our approach outperforms a state-of-the-art graph-search method in computation time and probability of reaching the goal.
- Open Access
- 30 October 2023
- Published by Springer Nature
- p. 3-24
XAI Requirements in Smart Production Processes: A Case Study
The increasing prevalence of artificial intelligence (AI) systems has led to a growing consensus on the importance of the explainability of such systems. This is often emphasized with respect to societal and developmental contexts, but it is also crucial within the context of business processes, including manufacturing and production. While this is widely recognized, there is a notable lack of practical examples that demonstrate how to take explainability into account in the latter contexts. This paper presents a real-world use case in which we employed AI to optimize an Industry 4.0 production process without considering explainable AI (XAI) requirements. Building on previous work on models of the relationship between XAI methods and various associated expectations, as well as non-functional explainability requirements, we show how business-oriented XAI requirements can be formulated and prepared for integration into process design. This case study is a valuable resource for researchers and practitioners seeking better to understand the role of explainable AI in practice.
- 30 October 2023
- Published by Springer Nature
- p. 3-24
Open AccessXAI Requirements in Smart Production Processes: A Case Study
The increasing prevalence of artificial intelligence (AI) systems has led to a growing consensus on the importance of the explainability of such systems. This is often emphasized with respect to societal and developmental contexts, but it is also crucial within the context of business processes, including manufacturing and production. While this is widely recognized, there is a notable lack of practical examples that demonstrate how to take explainability into account in the latter contexts. This paper presents a real-world use case in which we employed AI to optimize an Industry 4.0 production process without considering explainable AI (XAI) requirements. Building on previous work on models of the relationship between XAI methods and various associated expectations, as well as non-functional explainability requirements, we show how business-oriented XAI requirements can be formulated and prepared for integration into process design. This case study is a valuable resource for researchers and practitioners seeking better to understand the role of explainable AI in practice.
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Modeling and Computer Simulation
- Vol. 33 (4) , 1-28
DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning – Extended Version
Neural networks (NN) are gaining importance in sequential decision-making. Deep reinforcement learning (DRL), in particular, is extremely successful in learning action policies in complex and dynamic environments. Despite this success, however, DRL technology is not without its failures, especially in safety-critical applications: (i) the training objective maximizes average rewards, which may disregard rare but critical situations and hence lack local robustness; (ii) optimization objectives targeting safety typically yield degenerated reward structures, which, for DRL to work, must be replaced with proxy objectives. Here, we introduce a methodology that can help to address both deficiencies. We incorporate evaluation stages (ES) into DRL, leveraging recent work on deep statistical model checking (DSMC), which verifies NN policies in Markov decision processes. Our ES apply DSMC at regular intervals to determine state space regions with weak performance. We adapt the subsequent DRL training priorities based on the outcome, (i) focusing DRL on critical situations and (ii) allowing to foster arbitrary objectives. We run case studies on two benchmarks. One of them is the Racetrack, an abstraction of autonomous driving that requires navigating a map without crashing into a wall. The other is MiniGrid, a widely used benchmark in the AI community. Our results show that DSMC-based ES can significantly improve both (i) and (ii).
- 26 October 2023
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Modeling and Computer Simulation
- Vol. 33 (4) , 1-28
Open AccessDSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning – Extended Version
Neural networks (NN) are gaining importance in sequential decision-making. Deep reinforcement learning (DRL), in particular, is extremely successful in learning action policies in complex and dynamic environments. Despite this success, however, DRL technology is not without its failures, especially in safety-critical applications: (i) the training objective maximizes average rewards, which may disregard rare but critical situations and hence lack local robustness; (ii) optimization objectives targeting safety typically yield degenerated reward structures, which, for DRL to work, must be replaced with proxy objectives. Here, we introduce a methodology that can help to address both deficiencies. We incorporate evaluation stages (ES) into DRL, leveraging recent work on deep statistical model checking (DSMC), which verifies NN policies in Markov decision processes. Our ES apply DSMC at regular intervals to determine state space regions with weak performance. We adapt the subsequent DRL training priorities based on the outcome, (i) focusing DRL on critical situations and (ii) allowing to foster arbitrary objectives. We run case studies on two benchmarks. One of them is the Racetrack, an abstraction of autonomous driving that requires navigating a map without crashing into a wall. The other is MiniGrid, a widely used benchmark in the AI community. Our results show that DSMC-based ES can significantly improve both (i) and (ii).
- Open Access
- 13 December 2022
- Published by Springer Nature in International Journal on Software Tools for Technology Transfer
- Vol. 25 (3) , 407-426
Analyzing neural network behavior through deep statistical model checking
Neural networks (NN) are taking over ever more decisions thus far taken by humans, even though verifiable system-level guarantees are far out of reach. Neither is the verification technology available, nor is it even understood what a formal, meaningful, extensible, and scalable testbed might look like for such a technology. The present paper is an attempt to improve on both the above aspects. We present a family of formal models that contain basic features of automated decision-making contexts and which can be extended with further orthogonal features, ultimately encompassing the scope of autonomous driving. Due to the possibility to model random noise in the decision actuation, each model instance induces a Markov decision process (MDP) as verification object. The NN in this context has the duty to actuate (near-optimal) decisions. From the verification perspective, the externally learnt NN serves as a determinizer of the MDP, the result being a Markov chain which as such is amenable to statistical model checking. The combination of an MDP and an NN encoding the action policy is central to what we call “deep statistical model checking” (DSMC). While being a straightforward extension of statistical model checking, it enables to gain deep insight into questions like “how high is the NN-induced safety risk?”, “how good is the NN compared to the optimal policy?” (obtained by model checking the MDP), or “does further training improve the NN?”. We report on an implementation of DSMC inside the Modest Toolset in combination with externally learnt NNs, demonstrating the potential of DSMC on various instances of the model family, and illustrating its scalability as a function of instance size as well as other factors like the degree of NN training.
- 13 December 2022
- Published by Springer Nature in International Journal on Software Tools for Technology Transfer
- Vol. 25 (3) , 407-426
Open AccessAnalyzing neural network behavior through deep statistical model checking
Neural networks (NN) are taking over ever more decisions thus far taken by humans, even though verifiable system-level guarantees are far out of reach. Neither is the verification technology available, nor is it even understood what a formal, meaningful, extensible, and scalable testbed might look like for such a technology. The present paper is an attempt to improve on both the above aspects. We present a family of formal models that contain basic features of automated decision-making contexts and which can be extended with further orthogonal features, ultimately encompassing the scope of autonomous driving. Due to the possibility to model random noise in the decision actuation, each model instance induces a Markov decision process (MDP) as verification object. The NN in this context has the duty to actuate (near-optimal) decisions. From the verification perspective, the externally learnt NN serves as a determinizer of the MDP, the result being a Markov chain which as such is amenable to statistical model checking. The combination of an MDP and an NN encoding the action policy is central to what we call “deep statistical model checking” (DSMC). While being a straightforward extension of statistical model checking, it enables to gain deep insight into questions like “how high is the NN-induced safety risk?”, “how good is the NN compared to the optimal policy?” (obtained by model checking the MDP), or “does further training improve the NN?”. We report on an implementation of DSMC inside the Modest Toolset in combination with externally learnt NNs, demonstrating the potential of DSMC on various instances of the model family, and illustrating its scalability as a function of instance size as well as other factors like the degree of NN training.
- Open Access
- 6 August 2022
- Published by Springer Nature
- p. 430-443
MoGym: Using Formal Models for Training and Verifying Decision-making Agents
MoGym, is an integrated toolbox enabling the training and verification of machine-learned decision-making agents based on formal models, for the purpose of sound use in the real world. Given a formal representation of a decision-making problem in the JANI format and a reach-avoid objective, MoGym(a) enables training a decision-making agent with respect to that objective directly on the model using reinforcement learning (RL) techniques, and (b) it supports rigorous assessment of the quality of the induced decision-making agent by means of deep statistical model checking (DSMC). MoGymimplements the standard interface for training environments established by OpenAI Gym, thereby connecting to the vast body of existing work in the RL community. In return, it makes accessible the large set of existing JANI model checking benchmarks to machine learning research. It thereby contributes an efficient feedback mechanism for improving in particular reinforcement learning algorithms. The connective part is implemented on top of Momba. For the DSMC quality assurance of the learned decision-making agents, a variant of the statistical model checkermodesof the ModestToolsetis leveraged, which has been extended by two new resolution strategies for non-determinism when encountered during statistical evaluation.
- 6 August 2022
- Published by Springer Nature
- p. 430-443
Open AccessMoGym: Using Formal Models for Training and Verifying Decision-making Agents
MoGym, is an integrated toolbox enabling the training and verification of machine-learned decision-making agents based on formal models, for the purpose of sound use in the real world. Given a formal representation of a decision-making problem in the JANI format and a reach-avoid objective, MoGym(a) enables training a decision-making agent with respect to that objective directly on the model using reinforcement learning (RL) techniques, and (b) it supports rigorous assessment of the quality of the induced decision-making agent by means of deep statistical model checking (DSMC). MoGymimplements the standard interface for training environments established by OpenAI Gym, thereby connecting to the vast body of existing work in the RL community. In return, it makes accessible the large set of existing JANI model checking benchmarks to machine learning research. It thereby contributes an efficient feedback mechanism for improving in particular reinforcement learning algorithms. The connective part is implemented on top of Momba. For the DSMC quality assurance of the learned decision-making agents, a variant of the statistical model checkermodesof the ModestToolsetis leveraged, which has been extended by two new resolution strategies for non-determinism when encountered during statistical evaluation.
- 19 August 2021
- Published by Springer Nature
- p. C1
Correction to: DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning
- 19 August 2021
- Published by Springer Nature
- p. C1
Correction to: DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning
- 19 August 2021
- Published by Springer Nature
- p. 197-216
DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning
- 19 August 2021
- Published by Springer Nature
- p. 197-216
DSMC Evaluation Stages: Fostering Robust and Safe Behavior in Deep Reinforcement Learning
- 5 August 2021
- Published by Springer Nature
- p. 27-46
TraceVis: Towards Visualization for Deep Statistical Model Checking
- 5 August 2021
- Published by Springer Nature
- p. 27-46
TraceVis: Towards Visualization for Deep Statistical Model Checking