Robust Perception through Analysis by Synthesis

25 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Excessive Invariance Causes Adversarial Vulnerability (2018, methodology): “While classifying MNIST digits is considered a toy example, it has been shown to be challenging from the perspective of -robustness (Schott et al., 2018).”
VALUATION : A N ADVERSARIAL APPROACH TO UNCOVER CATASTROPHIC FAILURES (2018, unknown): “Numerous recent papers have questioned the practical value of norm ball robustness, when so many other forms of adversarial perturbations exist (Gilmer et al., 2018; Engstrom et al., 2017; Schott et al., 2018).”
On the Connection Between Adversarial Robustness and Saliency Map Interpretability (2019, background): “, 2016) as well as defenses against these (see (Schott et al., 2018) for an overview).”
Relating Adversarially Robust Generalization to Flat Minima (2021, methodology): “1, 3 [109] Lukas Schott, Jonas Rauber, Wieland Brendel, and Matthias Bethge.”
On Adaptive Attacks to Adversarial Example Defenses (2020, unknown): “Instead, it is better to employ a more diverse set of attacks, e.g. by including the score-based pixel-wise attack (Schott et al., 2019b) or the Brendel & Bethge (B&B) attack (Brendel et al., 2019) (the latter of which was not available at the time of initial publication).”