Search Results - "Könighofer, Bettina"
-
1
Online shielding for reinforcement learning
Published in Innovations in systems and software engineering (01-12-2023)“…Besides the recent impressive results on reinforcement learning (RL), safety is still one of the major research challenges in RL. RL is a machine-learning…”
Get full text
Journal Article -
2
Learning and Repair of Deep Reinforcement Learning Policies from Fuzz-Testing Data
Published in 2024 IEEE/ACM 46th International Conference on Software Engineering (ICSE) (14-04-2024)“…Reinforcement learning from demonstrations (RLfD) is a promising approach to improve the exploration efficiency of reinforcement learning (RL) by learning from…”
Get full text
Conference Proceeding -
3
Synthesizing robust systems
Published in Acta informatica (01-06-2014)“…Systems should not only be correct but also robust in the sense that they behave reasonably in unexpected situations. This article addresses synthesis of…”
Get full text
Journal Article Conference Proceeding -
4
Synthesis of Minimum-Cost Shields for Multi-agent Systems
Published in 2019 American Control Conference (ACC) (01-01-2019)“…In this paper, we propose a general approach to derive runtime enforcement implementations for multiagent systems, called shields, from temporal logical…”
Get full text
Conference Proceeding Journal Article -
5
Synthesizing Robust Systems with RATSY
Published in Electronic proceedings in theoretical computer science (03-07-2012)“…Specifications for reactive systems often consist of environment assumptions and system guarantees. An implementation should not only be correct, but also…”
Get full text
Journal Article -
6
Shield synthesis
Published in Formal methods in system design (2017)“…Shield synthesis is an approach to enforce safety properties at runtime. A shield monitors the system and corrects any erroneous output values instantaneously…”
Get full text
Journal Article -
7
Synthesis of synchronization using uninterpreted functions
Published in 2014 Formal Methods in Computer-Aided Design (FMCAD) (01-10-2014)“…Correctness of a program with respect to concurrency is often hard to achieve, but easy to specify: the concurrent program should produce the same results as a…”
Get full text
Conference Proceeding -
8
Test Where Decisions Matter: Importance-driven Testing for Deep Reinforcement Learning
Published 12-11-2024“…In many Deep Reinforcement Learning (RL) problems, decisions in a trained policy vary in significance for the expected safety and performance of the policy…”
Get full text
Journal Article -
9
Learning Environment Models with Continuous Stochastic Dynamics
Published 29-06-2023“…Solving control tasks in complex environments automatically through learning offers great potential. While contemporary techniques from deep reinforcement…”
Get full text
Journal Article -
10
Online Shielding for Reinforcement Learning
Published 04-12-2022“…Besides the recent impressive results on reinforcement learning (RL), safety is still one of the major research challenges in RL. RL is a machine-learning…”
Get full text
Journal Article -
11
Correct-by-Construction Runtime Enforcement in AI -- A Survey
Published 30-08-2022“…Runtime enforcement refers to the theories, techniques, and tools for enforcing correct behavior with respect to a formal specification of systems at runtime…”
Get full text
Journal Article -
12
Search-Based Testing of Reinforcement Learning
Published 07-05-2022“…Evaluation of deep reinforcement learning (RL) is inherently challenging. Especially the opaqueness of learned policies and the stochastic nature of both…”
Get full text
Journal Article -
13
Safety Shielding under Delayed Observation
Published 05-07-2023“…Agents operating in physical environments need to be able to handle delays in the input and output signals since neither data transmission nor sensing or…”
Get full text
Journal Article -
14
'Put the Car on the Stand': SMT-based Oracles for Investigating Decisions
Published 09-05-2023“…Principled accountability in the aftermath of harms is essential to the trustworthy design and governance of algorithmic decision making. Legal theory offers a…”
Get full text
Journal Article -
15
TEMPEST -- Synthesis Tool for Reactive Systems and Shields in Probabilistic Environments
Published 26-05-2021“…We present Tempest, a synthesis tool to automatically create correct-by-construction reactive systems and shields from qualitative or quantitative…”
Get full text
Journal Article -
16
Automata Learning meets Shielding
Published 04-12-2022“…Safety is still one of the major research challenges in reinforcement learning (RL). In this paper, we address the problem of how to avoid safety violations of…”
Get full text
Journal Article -
17
Online Shielding for Stochastic Systems
Published 17-12-2020“…In this paper, we propose a method to develop trustworthy reinforcement learning systems. To ensure safety especially during exploration, we automatically…”
Get full text
Journal Article -
18
Analyzing Intentional Behavior in Autonomous Agents under Uncertainty
Published 04-07-2023“…Principled accountability for autonomous decision-making in uncertain environments requires distinguishing intentional outcomes from negligent designs from…”
Get full text
Journal Article -
19
Formal Methods for Trused AI
Published in 2023 Formal Methods in Computer-Aided Design (FMCAD) (24-10-2023)“…The enormous influence of systems deploying AI is contrasted by the growing concerns about their safety and the relative lack of trust by the society. This…”
Get full text
Conference Proceeding -
20
Synthesis of Admissible Shields
Published 15-04-2019“…Hardware and Software: Verification and Testing - 12th International Haifa Verification Conference, {HVC} 2016, Haifa, Israel, November 14-17, 2016,…”
Get full text
Journal Article