Roland S. Zimmermann

neural network (2) generative model (2) adversarial robustness (1) independent component analysis (1) self-supervised learning (1) neural network interpretability (1) model analysis (1) theoretical analysis (1) convolutional neural network (1) feature visualization (1) mechanistic interpretability (1) model scaling (1) adversarial example (1) object-centric representation (1) vision model (1) robustness evaluation (1) explainable artificial intelligence (1) attack evaluation (1) representational identifiability (1) defense evaluation (1)