Decision-making under uncertainty: beyond probabilities: Challenges and perspectives
This position paper reflects on the state-of-the-art in decision-making under uncertainty. A
classical assumption is that probabilities can sufficiently capture all uncertainty in a system …
classical assumption is that probabilities can sufficiently capture all uncertainty in a system …
Robust control for dynamical systems with non-gaussian noise via formal abstractions
Controllers for dynamical systems that operate in safety-critical settings must account for
stochastic disturbances. Such disturbances are often modeled as process noise in a …
stochastic disturbances. Such disturbances are often modeled as process noise in a …
Robust anytime learning of Markov decision processes
Markov decision processes (MDPs) are formal models commonly used in sequential
decision-making. MDPs capture the stochasticity that may arise, for instance, from imprecise …
decision-making. MDPs capture the stochasticity that may arise, for instance, from imprecise …
Multi-agent reinforcement learning with temporal logic specifications
In this paper, we study the problem of learning to satisfy temporal logic specifications with a
group of agents in an unknown environment, which may exhibit probabilistic behaviour …
group of agents in an unknown environment, which may exhibit probabilistic behaviour …
Optimistic value iteration
A Hartmanns, BL Kaminski - International Conference on Computer Aided …, 2020 - Springer
Markov decision processes are widely used for planning and verification in settings that
combine controllable or adversarial choices with probabilistic behaviour. The standard …
combine controllable or adversarial choices with probabilistic behaviour. The standard …
A framework for transforming specifications in reinforcement learning
Reactive synthesis algorithms allow automatic construction of policies to control an
environment modeled as a Markov Decision Process (MDP) that are optimal with respect to …
environment modeled as a Markov Decision Process (MDP) that are optimal with respect to …
Policy synthesis and reinforcement learning for discounted LTL
The difficulty of manually specifying reward functions has led to an interest in using linear
temporal logic (LTL) to express objectives for reinforcement learning (RL). However, LTL …
temporal logic (LTL) to express objectives for reinforcement learning (RL). However, LTL …
Constructing MDP abstractions using data with formal guarantees
This letter is concerned with a data-driven technique for constructing finite Markov decision
processes (MDPs) as finite abstractions of discrete-time stochastic control systems with …
processes (MDPs) as finite abstractions of discrete-time stochastic control systems with …
On correctness, precision, and performance in quantitative verification: QComp 2020 competition report
Quantitative verification tools compute probabilities, expected rewards, or steady-state
values for formal models of stochastic and timed systems. Exact results often cannot be …
values for formal models of stochastic and timed systems. Exact results often cannot be …
A PAC learning algorithm for LTL and omega-regular objectives in MDPs
Linear temporal logic (LTL) and omega-regular objectives---a superset of LTL---have seen
recent use as a way to express non-Markovian objectives in reinforcement learning. We …
recent use as a way to express non-Markovian objectives in reinforcement learning. We …