101 citations · 513 across the 16 of their papers we have counts for
Showing 2017Show all
2 papers · 1 filter
cs.LG2017★ 82 cited
Certified Defenses for Data Poisoning Attacks
Jacob Steinhardt, Pang Wei Koh, Percy Liang
Machine learning systems trained on user-provided data are susceptible to data poisoning attacks, whereby malicious users inject false training data with the aim of corrupting the…
stat.ML2017
Understanding Black-box Predictions via Influence Functions
Pang Wei Koh, Percy Liang
How can we explain the predictions of a black-box model? In this paper, we use influence functions -- a classic technique from robust statistics -- to trace a model's prediction th…