Robert Geirhos
Robert Geirhos
Research Scientist, Google DeepMind
Bestätigte E-Mail-Adresse bei - Startseite
Zitiert von
Zitiert von
ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness
R Geirhos, P Rubisch, C Michaelis, M Bethge, FA Wichmann, W Brendel
Oral @ International Conference on Learning Representations (ICLR 2019), 2019
Shortcut Learning in Deep Neural Networks
R Geirhos, JH Jacobsen, C Michaelis, R Zemel, W Brendel, M Bethge, ...
Nature Machine Intelligence 2 (11), 665-673, 2020
Generalisation in humans and deep neural networks
R Geirhos, CR Medina Temme, J Rauber, HH Schütt, M Bethge, ...
Advances in Neural Information Processing Systems 31 (NeurIPS 2018), 2018
Benchmarking Robustness in Object Detection: Autonomous Driving when Winter is Coming
C Michaelis, B Mitzkus, R Geirhos, E Rusak, O Bringmann, AS Ecker, ...
Machine Learning for Autonomous Driving Workshop (NeurIPS 2019), 2019
Comparing deep neural networks against humans: object recognition when the signal gets weaker
R Geirhos, DHJ Janssen, HH Schütt, J Rauber, M Bethge, FA Wichmann
arXiv preprint arXiv:1706.06969, 2017
Scaling vision transformers to 22 billion parameters
M Dehghani, J Djolonga, B Mustafa, P Padlewski, J Heek, J Gilmer, ...
Oral @ International Conference on Machine Learning (ICML 2023), 2023
Beyond neural scaling laws: beating power law scaling via data pruning
B Sorscher, R Geirhos, S Shekhar, S Ganguli, AS Morcos
Outstanding Paper Award @ Advances in Neural Information Processing Systems …, 2022
Partial success in closing the gap between human and machine vision
R Geirhos, K Narayanappa, B Mitzkus, T Thieringer, M Bethge, ...
Oral @ Advances in Neural Information Processing Systems 34 (NeurIPS 2021), 2021
Beyond accuracy: quantifying trial-by-trial behaviour of CNNs and humans by measuring error consistency
R Geirhos, K Meding, FA Wichmann
Advances in Neural Information Processing Systems 33 (NeurIPS 2020), 2020
On the surprising similarities between supervised and self-supervised models
R Geirhos, K Narayanappa, B Mitzkus, M Bethge, FA Wichmann, ...
Oral @ Shared Visual Representations in Humans & Machines Workshop (NeurIPS …, 2020
Exemplary Natural Images Explain CNN Activations Better than State-of-the-Art Feature Visualization
J Borowski, RS Zimmermann, J Schepers, R Geirhos, TSA Wallis, ...
International Conference on Learning Representations (ICLR 2021), 2020
Trivial or impossible--dichotomous data difficulty masks model differences (on ImageNet and beyond)
K Meding, LMS Buschoff, R Geirhos, FA Wichmann
International Conference on Learning Representations (ICLR 2022), 2021
Are deep neural networks adequate behavioral models of human visual perception?
FA Wichmann, R Geirhos
Annual Review of Vision Science 9, 501-524, 2023
How Well do Feature Visualizations Support Causal Understanding of CNN Activations?
RS Zimmermann, J Borowski, R Geirhos, M Bethge, TSA Wallis, ...
Spotlight @ Advances in Neural Information Processing Systems 34 (NeurIPS 2021), 2021
The developmental trajectory of object recognition robustness: Children are like small adults but unlike big deep neural networks
LS Huber, R Geirhos, FA Wichmann
Journal of Vision 23 (7), 4-4, 2023
Methods and measurements to compare men against machines
FA Wichmann, DHJ Janssen, R Geirhos, G Aguilar, HH Schütt, ...
Electronic Imaging 29, 36-45, 2017
Getting aligned on representational alignment
I Sucholutsky, L Muttenthaler, A Weller, A Peng, A Bobu, B Kim, BC Love, ...
arXiv preprint arXiv:2310.13018, 2023
Patch n'Pack: NaViT, a Vision Transformer for any Aspect Ratio and Resolution
M Dehghani, B Mustafa, J Djolonga, J Heek, M Minderer, M Caron, ...
Advances in Neural Information Processing Systems 36 (NeurIPS 2023), 2023
Intriguing properties of generative classifiers
P Jaini, K Clark, R Geirhos
Spotlight @ International Conference on Learning Representations (ICLR 2024), 2023
Don't trust your eyes: on the (un) reliability of feature visualizations
R Geirhos, RS Zimmermann, B Bilodeau, W Brendel, B Kim
New Frontiers in Adversarial Machine Learning Workshop (ICML 2023), 2023
Das System kann den Vorgang jetzt nicht ausführen. Versuchen Sie es später erneut.
Artikel 1–20