no code implementations • 11 Aug 2023 • Hendrik Vogt, Stefan Buehler, Mark Schutera
To address this, we propose to encapsulate the neural network under deployment within an uncertainty estimation envelope that is based on the epistemic uncertainty estimation through the Monte Carlo Dropout approach.
1 code implementation • PLOS ONE 2022 • Mark Schutera, Luca Rettenberger, Christian Pylatiuk, Markus Reischl
This paper discusses the effects of frugal labeling and proposes to train neural networks for multi-class semantic segmentation on heterogeneously labeled data based on a novel objective function.
Ranked #1 on 2D Semantic Segmentation on Extended heartSeg
no code implementations • 27 Nov 2018 • Mark Schutera, Niklas Goby, Stefan Smolarek, Markus Reischl
Different strategies, containing fixed-time, gap-based and time-based control approaches as well as our deep reinforcement learning based control approach, are implemented and assessed.
no code implementations • 19 Oct 2018 • Mark Schutera, Niklas Goby, Dirk Neumann, Markus Reischl
Transportation and traffic are currently undergoing a rapid increase in terms of both scale and complexity.