Probabilistic inference and o.o.d. or few-shot generalization benchmarking

  • A note on the evaluation of generative models (L Theis, A Oord, M Bethge)
  • Reliable attacks against black-box machine learning models (W Brendel, J Rauber, M Bethge)
  • Shortcut learning in deep neural networks (R Geirhos, JH Jacobsen, C Michaelis, R Zemel, W Brendel, M Bethge, …)
  • Foolbox: A python toolbox to benchmark the robustness of machine learning models (J Rauber, W Brendel, M Bethge)
  • Generalisation in humans and deep neural networks (R Geirhos, CRM Temme, J Rauber, HH Schütt, M Bethge, FA Wichmann)
  • Information-theoretic model comparison unifies saliency metrics (M Kümmerer, TSA Wallis, M Bethge)
  • Saliency benchmarking made easy: Separating models, maps and metrics (M Kümmerer, TSA Wallis, M Bethge)
  • Benchmarking spike rate inference in population calcium imaging (L Theis, P Berens, E Froudarakis, J Reimer, MR Rosón, T Baden, T Euler, …)
  • Benchmarking robustness in object detection: Autonomous driving when winter is coming (C Michaelis, B Mitzkus, R Geirhos, E Rusak, O Bringmann, AS Ecker, …)
  • If your data distribution shifts, use self-learning (E Rusak, S Schneider, G Pachitariu, L Eck, PV Gehler, O Bringmann, W Brendel, M Bethge)
  • ImageNet-D: A new challenging robustness dataset inspired by domain adaptation (E Rusak, S Schneider, PV Gehler, O Bringmann, W Brendel, M Bethge)
  • Benchmarking Unsupervised Object Representations for Video Sequences (MA Weis, K Chitta, Y Sharma, W Brendel, M Bethge, A Geiger, AS Ecker)
  • In All Likelihood, Deep Belief Is Not Enough (L Theis, S Gerwinn, F Sinz, M Bethge)
  • Engineering a less artificial intelligence (FH Sinz, X Pitkow, J Reimer, M Bethge, AS Tolias)
  • Foolbox native: Fast adversarial attacks to benchmark the robustness of machine learning models in pytorch, tensorflow, and jax (J Rauber, R Zimmermann, M Bethge, W Brendel)
  • Accurate, reliable and fast robustness evaluation (W Brendel, J Rauber, M Kümmerer, I Ustyuzhaninov, M Bethge)
  • Factorial coding of natural images: how effective are linear models in removing higher-order dependencies? (M Bethge)
  • Partial success in closing the gap between human and machine vision (R Geirhos, K Narayanappa, B Mitzkus, T Thieringer, M Bethge, …)
  • Measuring the importance of temporal features in video saliency (M Tangemann, M Kümmerer, TSA Wallis, M Bethge)
  • State-of-the-Art in Human Scanpath Prediction (M Kümmerer, M Bethge)
Matthias Bethge
Matthias Bethge
Professor for Computational Neuroscience and Machine Learning & Director of the Tübingen AI Center

Matthias Bethge is Professor for Computational Neuroscience and Machine Learning at the University of Tübingen and director of the Tübingen AI Center, a joint center between Tübingen University and MPI for Intelligent Systems that is part of the German AI strategy.