283 citations · 1k across the 14 of their papers we have counts for
Showing 2018Show all
2 papers · 1 filter
stat.ML2018★ 27 cited
Sufficient Conditions for Idealised Models to Have No Adversarial Examples: a Theoretical and Empirical Study with Bayesian Neural Networks
Yarin Gal, Lewis Smith
We prove, under two sufficient conditions, that idealised models can have no adversarial examples. We discuss which idealised models satisfy our conditions, and show that idealised…
stat.ML2018★ 176 cited
Understanding Measures of Uncertainty for Adversarial Example Detection
Lewis Smith, Yarin Gal
Measuring uncertainty is a promising technique for detecting adversarial examples, crafted inputs on which the model predicts an incorrect class with high confidence. But many meas…