Zobrazeno 1 - 10
of 1 637
pro vyhledávání: '"A, Boetius"'
Probabilistic verification of neural networks is concerned with formally analysing the output distribution of a neural network under a probability distribution of the inputs. Examples of probabilistic verification include verifying the demographic pa
Externí odkaz:
http://arxiv.org/abs/2405.17556
Autor:
Boetius, David, Leue, Stefan
Naively trained Deep Reinforcement Learning agents may fail to satisfy vital safety constraints. To avoid costly retraining, we may desire to repair a previously trained reinforcement learning agent to obviate unsafe behaviour. We devise a counterexa
Externí odkaz:
http://arxiv.org/abs/2405.15430
Autor:
Boetius, David, Leue, Stefan
Current approaches to neural network verification focus on specifications that target small regions around known input data points, such as local robustness. Thus, using these approaches, we can not obtain guarantees for inputs that are not close to
Externí odkaz:
http://arxiv.org/abs/2306.12495
Counterexample-guided repair aims at creating neural networks with mathematical safety guarantees, facilitating the application of neural networks in safety-critical domains. However, whether counterexample-guided repair is guaranteed to terminate re
Externí odkaz:
http://arxiv.org/abs/2301.11342
Autor:
Gunter Wegener, Massimiliano Molari, Autun Purser, Alexander Diehl, Elmar Albers, Maren Walter, Christian Mertens, Christopher R. German, Antje Boetius
Publikováno v:
Frontiers in Microbiology, Vol 15 (2024)
Hydrothermal vents emit hot fluids enriched in energy sources for microbial life. Here, we compare the ecological and biogeochemical effects of hydrothermal venting of two recently discovered volcanic seamounts, Polaris and Aurora of the Gakkel Ridge
Externí odkaz:
https://doaj.org/article/96ac6b0d96cc47c6a44908bd240f1774
Autor:
Albers, Elmar, Behrendt, Nele, Diehl, Alexander, Genske, Felix, Monien, Patrick, Kasemann, Simone A., Purser, Autun, Boetius, Antje, Bach, Wolfgang
Publikováno v:
In Marine Geology September 2024 475
Publikováno v:
SPIN 2022
Deep neural networks (DNNs) are increasingly applied in safety-critical domains, such as self-driving cars, unmanned aircraft, and medical diagnosis. It is of fundamental importance to certify the safety of these DNNs, i.e. that they comply with a fo
Externí odkaz:
http://arxiv.org/abs/2106.01917
Autor:
Christopher R. German, Eoghan P. Reeves, Andreas Türke, Alexander Diehl, Elmar Albers, Wolfgang Bach, Autun Purser, Sofia P. Ramalho, Stefano Suman, Christian Mertens, Maren Walter, Eva Ramirez-Llodra, Vera Schlindwein, Stefan Bünz, Antje Boetius
Publikováno v:
Nature Communications, Vol 13, Iss 1, Pp 1-11 (2022)
The Aurora hydrothermal field (Arctic Ocean) is hosted in volcanic rocks but also shows evidence of mantle rock influence in the shallow sub-surface. Our discovery is pertinent to disciplines from marine mining to the search for life beyond Earth.
Externí odkaz:
https://doaj.org/article/07ec42a41af743b49eff0e327c9b7189
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Autor:
Magda G. Cardozo-Mino, Ian Salter, Eva-Maria Nöthig, Katja Metfies, Simon Ramondenc, Claudia Wekerle, Thomas Krumpen, Antje Boetius, Christina Bienhold
Publikováno v:
Frontiers in Marine Science, Vol 10 (2023)
Marine sinking particles sequester atmospheric carbon dioxide to the deep ocean via the biological carbon pump. Understanding how environmental shifts drive changes in the microbial composition of particles, and how these affect the export of organic
Externí odkaz:
https://doaj.org/article/2b53f2d1dd574df49c649df0767759f3