2.3k citations · 4.9k across the 24 of their papers we have counts for
Showing stat.MLShow all
2 papers · 1 filter
stat.ML2018★ 96 cited
Stronger Data Poisoning Attacks Break Data Sanitization Defenses
Pang Wei Koh, Jacob Steinhardt, Percy Liang
Machine learning models trained on data from the outside world can be corrupted by data poisoning attacks that inject malicious points into the models' training sets. A common defe…
stat.ML2017★ 1.3k cited
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…