Roland S. Zimmermann

发表

Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations pdf

Matthias Bethge, Robert Geirhos, Wieland Brendel, 2020, ArXiv.

A Simple Way to Make Neural Networks Robust Against Diverse Image Corruptions

Matthias Bethge, Oliver Bringmann, Lukas Schott, 2020, ECCV.

Observing spatio-temporal dynamics of excitable media using reservoir computing.

Ulrich Parlitz, Roland S. Zimmermann, U. Parlitz, 2018, Chaos.

Foolbox Native: Fast adversarial attacks to benchmark the robustness of machine learning models in PyTorch, TensorFlow, and JAX

Matthias Bethge, Wieland Brendel, Jonas Rauber, 2020, J. Open Source Softw..

Natural images are more informative for interpreting CNN activations than synthetic feature visualizations

Matthias Bethge, Robert Geirhos, Wieland Brendel, 2020 .

Comment on "Adv-BNN: Improved Adversarial Defense through Robust Bayesian Neural Network" pdf

Roland S. Zimmermann, 2019, ArXiv.

Faster Training of Mask R-CNN by Focusing on Instance Boundaries

Roland S. Zimmermann, Julien N. Siems, 2018, Comput. Vis. Image Underst..

Simion Zoo: A Workbench for Distributed Experimentation with Reinforcement Learning for Continuous Control Tasks pdf

Manuel Graña, Borja Fernández-Gauna, Roland S. Zimmermann, 2019, ArXiv.

How Well do Feature Visualizations Support Causal Understanding of CNN Activations? pdf

Matthias Bethge, Robert Geirhos, Wieland Brendel, 2021, NeurIPS.

Score-Based Generative Classifiers pdf

Lukas Schott, Roland S. Zimmermann, David A. Klindt, 2021, ArXiv.

Contrastive Learning Inverts the Data Generating Process pdf

Matthias Bethge, Wieland Brendel, Yash Sharma, 2021, ICML.

Natural Images are More Informative for Interpreting CNN Activations than State-of-the-Art Synthetic Feature Visualizations

Thomas S. A. Wallis, Roland S. Zimmermann, M. Bethge, 2020 .

A Simple Way to Make Neural Networks Robust Against Diverse Image Corruptions

Matthias Bethge, Oliver Bringmann, Lukas Schott, 2020, ECCV.

Increasing Confidence in Adversarial Robustness Evaluations

Roland S. Zimmermann, Florian Tramèr, Nicholas Carlini, 2022, NeurIPS.

Contrastive Learning Inverts the Data Generating Process

Roland S. Zimmermann, M. Bethge, Yash Sharma, 2021, ICML.

How Well do Feature Visualizations Support Causal Understanding of CNN Activations?

Thomas S. A. Wallis, Roland S. Zimmermann, M. Bethge, 2021, NeurIPS.

Exemplary Natural Images Explain CNN Activations Better than Feature Visualizations

Thomas S. A. Wallis, Roland S. Zimmermann, M. Bethge, 2020, arXiv.org.

Reconstructing Complex Cardiac Excitation Waves From Incomplete Data Using Echo State Networks and Convolutional Autoencoders

Roland S. Zimmermann, S. Luther, U. Parlitz, 2021, Frontiers in Applied Mathematics and Statistics.

Smart Grid for Industry Using Multi-Agent Reinforcement Learning

Andrea Hohmann, Gunther Reinhart, Martin Roesch, 2020, Applied Sciences.

Sensitivity of Slot-Based Object-Centric Models to their Number of Slots

Sjoerd van Steenkiste, Mehdi S. M. Sajjadi, Roland S. Zimmermann, 2023, ArXiv.

Provably Learning Object-Centric Representations

Roland S. Zimmermann, Julius von Kügelgen, Yash Sharma, 2023, ICML.

Increasing the robustness of DNNs against image corruptions by playing the Game of Noise pdf

Matthias Bethge, Oliver Bringmann, Lukas Schott, 2020, ArXiv.

I NCREASING THE ROBUSTNESS OF DNN S AGAINST IMAGE CORRUPTIONS BY PLAYING THE G AME OF N OISE

Roland S. Zimmermann, M. Bethge, Lukas Schott, 2020 .

Don't trust your eyes: on the (un)reliability of feature visualizations

Roland S. Zimmermann, Wieland Brendel, Been Kim, 2023, arXiv.org.

A Self-Supervised Feature Map Augmentation (FMA) Loss and Combined Augmentations Finetuning to Efficiently Improve the Robustness of CNNs

Chun Yuan, Peter Schlicht, Roland Zimmermann, 2020, CSCS.

Exemplary Natural Images Explain CNN Activations Better than State-of-the-Art Feature Visualization

Matthias Bethge, Robert Geirhos, Wieland Brendel, 2021, ICLR.

Scale Alone Does not Improve Mechanistic Interpretability in Vision Models

Roland S. Zimmermann, Wieland Brendel, Thomas Klein, 2023, NeurIPS.

Don't trust your eyes: on the (un)reliability of feature visualizations

Roland S. Zimmermann, Wieland Brendel, Been Kim, 2023, ArXiv.

Observing spatio-temporal dynamics of excitable media using reservoir computing.

Roland S. Zimmermann, U. Parlitz, 2018, Chaos.

InfoNCE: Identifying the Gap Between Theory and Practice

Roland S. Zimmermann, Wieland Brendel, E. Rusak, 2024, ArXiv.