Probabilistic safety guarantees for Markov decision processes
R Wisniewski, ML Bujorianu - IEEE Transactions on Automatic …, 2023 - ieeexplore.ieee.org
This article aims to incorporate safety specifications into Markov decision processes.
Explicitly, we address the minimization problem up to a stopping time with safety constraints …
Explicitly, we address the minimization problem up to a stopping time with safety constraints …
Safe reinforcement learning for constrained markov decision processes with stochastic stopping time
In this paper, we present an online reinforcement learning algorithm for constrained Markov
decision processes with a safety constraint. Despite the necessary attention of the scientific …
decision processes with a safety constraint. Despite the necessary attention of the scientific …
Online learning of safety function for Markov decision processes
A Mazumdar, R Wisniewski… - 2023 European Control …, 2023 - ieeexplore.ieee.org
In this paper, we aim to study safety specifications for a Markov decision process with
stochastic stopping time in an almost model-free setting. Our approach involves …
stochastic stopping time in an almost model-free setting. Our approach involves …
Stochastic safety for random dynamical systems
ML Bujorianu, R Wisniewski… - 2021 American Control …, 2021 - ieeexplore.ieee.org
In the paper, we study the so-called p-safety of a random dynamical system. We generalize
the existing results for safety barrier certificates for deterministic dynamical systems and …
the existing results for safety barrier certificates for deterministic dynamical systems and …
Stochastic Safety of Hybrid Markov Chains
In this paper we study the stochastic safety problem for a class of Markov chains generated
by iterated function systems (IFS). These Markov chains represent a mathematical construct …
by iterated function systems (IFS). These Markov chains represent a mathematical construct …
On Bellman's principle of optimality and Reinforcement learning for safety-constrained Markov decision process
We study optimality for the safety-constrained Markov decision process which is the
underlying framework for safe reinforcement learning. Specifically, we consider a …
underlying framework for safe reinforcement learning. Specifically, we consider a …
Safe Dynamic Programming
R Wisniewski, ML Bujorianu - arXiv preprint arXiv:2109.03307, 2021 - arxiv.org
We incorporate safety specifications into dynamic programming. Explicitly, we address the
minimization problem of a Markov decision process up to a stopping time with safety …
minimization problem of a Markov decision process up to a stopping time with safety …
From MDP to POMDP and Back: Safety and Compositionality
ML Bujorianu, T Caulfield, D Pym… - 2023 European …, 2023 - ieeexplore.ieee.org
We propose a compositional framework for the stochastic safety of distributed Markov
Decision Processes (MDPs) and Partially Observable Markov Decision Processes …
Decision Processes (MDPs) and Partially Observable Markov Decision Processes …
Cyber-Physical Ecosystems: Modelling and Verification
ML Bujorianu - International Conference on Engineering of Computer …, 2023 - Springer
In this paper, we set up a mathematical framework for the modelling and verification of
complex cyber-physical ecosystems. In our setting, cyber-physical ecosystems are cyber …
complex cyber-physical ecosystems. In our setting, cyber-physical ecosystems are cyber …