1 code implementation • 28 Aug 2023 • Asger Horn Brorholt, Peter Gjøl Jensen, Kim Guldstrand Larsen, Florian Lorber, Christian Schilling
We experimentally demonstrate superiority of the pre-shielding approach.
no code implementations • 30 Jun 2020 • Roderick Bloem, Peter Gjøl Jensen, Bettina Könighofer, Kim Guldstrand Larsen, Florian Lorber, Alexander Palmisano
A timed pre-shield is placed before the system and provides a set of safe outputs.
no code implementations • 26 Jun 2020 • Manfred Jaeger, Giorgio Bacci, Giovanni Bacci, Kim Guldstrand Larsen, Peter Gjøl Jensen
Second, we use imprecise Markov decision process approximations as a tool to analyse and validate cost functions and strategies obtained by reinforcement learning.