no code implementations • 14 Feb 2024 • Leo Schwinn, David Dobre, Sophie Xhonneux, Gauthier Gidel, Stephan Gunnemann
We address this research gap and propose the embedding space attack, which directly attacks the continuous embedding representation of input tokens.
1 code implementation • 30 Oct 2023 • Leo Schwinn, David Dobre, Stephan Günnemann, Gauthier Gidel
Here, one major impediment has been the overestimation of the robustness of new defense approaches due to faulty defense evaluations.
no code implementations • 21 May 2023 • Dario Zanca, Andrea Zugarini, Simon Dietz, Thomas R. Altstidl, Mark A. Turban Ndjeuha, Leo Schwinn, Bjoern Eskofier
Understanding the mechanisms underlying human attention is a fundamental challenge for both vision science and artificial intelligence.
Ranked #1 on Scanpath prediction on CapMIT1003
no code implementations • 17 May 2023 • Thomas Altstidl, David Dobre, Björn Eskofier, Gauthier Gidel, Leo Schwinn
In this work, we demonstrate that a similar approach can substantially improve deterministic certified defenses.
2 code implementations • 3 May 2023 • Kai Klede, Leo Schwinn, Dario Zanca, Björn Eskofier
Clustering is at the very core of machine learning, and its applications proliferate with the increasing availability of data.
no code implementations • 22 Nov 2022 • Leo Schwinn, Doina Precup, Bjoern Eskofier, Dario Zanca
Existing models of human visual attention are generally unable to incorporate direct task guidance and therefore cannot model an intent or goal when exploring a scene.
1 code implementation • 18 Nov 2022 • Thomas Altstidl, An Nguyen, Leo Schwinn, Franz Köferl, Christopher Mutschler, Björn Eskofier, Dario Zanca
We also demonstrate that our family of models is able to generalize well towards larger scales and improve scale equivariance.
no code implementations • 19 May 2022 • Leo Schwinn, Leon Bungert, An Nguyen, René Raab, Falk Pulsmeyer, Doina Precup, Björn Eskofier, Dario Zanca
The reliability of neural networks is essential for their use in safety-critical applications.
no code implementations • 19 Apr 2022 • Leo Schwinn, Doina Precup, Björn Eskofier, Dario Zanca
By and large, existing computational models of visual attention tacitly assume perfect vision and full access to the stimulus and thereby deviate from foveated biological vision.
no code implementations • 21 May 2021 • Leo Schwinn, René Raab, An Nguyen, Dario Zanca, Bjoern Eskofier
Progress in making neural networks more robust against adversarial attacks is mostly marginal, despite the great efforts of the research community.
1 code implementation • 23 Mar 2021 • Leon Bungert, René Raab, Tim Roith, Leo Schwinn, Daniel Tenbrinck
Despite the large success of deep neural networks (DNN) in recent years, most neural networks still lack mathematical guarantees in terms of stability.
1 code implementation • 24 Feb 2021 • Leo Schwinn, An Nguyen, René Raab, Leon Bungert, Daniel Tenbrinck, Dario Zanca, Martin Burger, Bjoern Eskofier
The susceptibility of deep neural networks to untrustworthy predictions, including out-of-distribution (OOD) data and adversarial examples, still prevent their widespread use in safety-critical applications.
1 code implementation • 11 Jan 2021 • An Nguyen, Stefan Foerstel, Thomas Kittler, Andrey Kurzyukov, Leo Schwinn, Dario Zanca, Tobias Hipp, Da Jun Sun, Michael Schrapp, Eva Rothgang, Bjoern Eskofier
The overall framework is currently deployed, learns and evaluates predictive models from terabytes of IoT and enterprise data to actively monitor the customer sentiment for a fleet of thousands of high-end medical devices.
no code implementations • 5 Nov 2020 • Leo Schwinn, An Nguyen, René Raab, Dario Zanca, Bjoern Eskofier, Daniel Tenbrinck, Martin Burger
We empirically show that by incorporating this nonlocal gradient information, we are able to give a more accurate estimation of the global descent direction on noisy and non-convex loss surfaces.
1 code implementation • 21 Oct 2020 • An Nguyen, Wenyu Zhang, Leo Schwinn, Bjoern Eskofier
Process Mining has recently gained popularity in healthcare due to its potential to provide a transparent, objective and data-based view on processes.
1 code implementation • 2 Oct 2020 • An Nguyen, Srijeet Chatterjee, Sven Weinzierl, Leo Schwinn, Martin Matzner, Bjoern Eskofier
To better model the time dependencies between events, we propose a new PBPM technique based on time-aware LSTM (T-LSTM) cells.
no code implementations • 24 Feb 2020 • Leo Schwinn, René Raab, Björn Eskofier
Further, we add a learnable regularization step prior to the neural network, which we call Pixelwise Noise Injection Layer (PNIL).