Timezone: »

Certified Data Removal from Machine Learning Models
Chuan Guo · Tom Goldstein · Awni Hannun · Laurens van der Maaten

Thu Jul 16 08:00 AM -- 08:45 AM & Thu Jul 16 07:00 PM -- 07:45 PM (PDT) @ None #None

Good data stewardship requires removal of data at the request of the data's owner. This raises the question if and how a trained machine-learning model, which implicitly stores information about its training data, should be affected by such a removal request. Is it possible to ``remove'' data from a machine-learning model? We study this problem by defining certified removal: a very strong theoretical guarantee that a model from which data is removed cannot be distinguished from a model that never observed the data to begin with. We develop a certified-removal mechanism for linear classifiers and empirically study learning settings in which this mechanism is practical.

Author Information

Chuan Guo (Cornell University)
Tom Goldstein (University of Maryland)
Awni Hannun (Facebook AI Research)
Laurens van der Maaten (Facebook)

More from the Same Authors