Source author record

Martin Wirsing

Martin Wirsing appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Catalog footprint

What is connected

3works
4topics
4close collaborators

Actions

Connect this record

Log in to claim

Research graph

See the researcher in context

Open full explorer

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

3 published item(s)

preprint2023arXiv

A modular bisimulation characterisation for fragments of hybrid logic

There are known characterisations of several fragments of hybrid logic by means of invariance under bisimulations of some kind. The fragments include $\{\store, \jump\}$ with or without nominals (Areces, Blackburn, Marx), $\jump$ with or without nominals (ten Cate), and $\store$ without nominals (Hodkinson, Tahiri). Some pairs of these characterisations, however, are incompatible with one another. For other fragments of hybrid logic no such characterisations were known so far. We prove a generic bisimulation characterisation theorem for all standard fragments of hybrid logic, in particular for the case with $\store$ and nominals, left open by Hodkinson and Tahiri. Our characterisation is built on a common base and for each feature extension adds a specific condition, so it is modular in an engineering sense.

preprint2021arXiv

Synthesizing Safe Policies under Probabilistic Constraints with Reinforcement Learning and Bayesian Model Checking

We propose to leverage epistemic uncertainty about constraint satisfaction of a reinforcement learner in safety critical domains. We introduce a framework for specification of requirements for reinforcement learners in constrained settings, including confidence about results. We show that an agent's confidence in constraint satisfaction provides a useful signal for balancing optimization and safety in the learning process.

preprint2010arXiv

Modeling and Analyzing Adaptive User-Centric Systems in Real-Time Maude

Pervasive user-centric applications are systems which are meant to sense the presence, mood, and intentions of users in order to optimize user comfort and performance. Building such applications requires not only state-of-the art techniques from artificial intelligence but also sound software engineering methods for facilitating modular design, runtime adaptation and verification of critical system requirements. In this paper we focus on high-level design and analysis, and use the algebraic rewriting language Real-Time Maude for specifying applications in a real-time setting. We propose a generic component-based approach for modeling pervasive user-centric systems and we show how to analyze and prove crucial properties of the system architecture through model checking and simulation. For proving time-dependent properties we use Metric Temporal Logic (MTL) and present analysis algorithms for model checking two subclasses of MTL formulas: time-bounded response and time-bounded safety MTL formulas. The underlying idea is to extend the Real-Time Maude model with suitable clocks, to transform the MTL formulas into LTL formulas over the extended specification, and then to use the LTL model checker of Maude. It is shown that these analyses are sound and complete for maximal time sampling. The approach is illustrated by a simple adaptive advertising scenario in which an adaptive advertisement display can react to actions of the users in front of the display.