Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observations
From MaRDI portal
Publication:6336222
arXiv2003.02960MaRDI QIDQ6336222
Stefano Soatto, Alessandro Achille, Aditya Golatkar
Publication date: 5 March 2020
Abstract: We describe a procedure for removing dependency on a cohort of training data from a trained deep network that improves upon and generalizes previous methods to different readout functions and can be extended to ensure forgetting in the activations of the network. We introduce a new bound on how much information can be extracted per query about the forgotten cohort from a black-box network for which only the input-output behavior is observed. The proposed forgetting procedure has a deterministic part derived from the differential equations of a linearized version of the model, and a stochastic part that ensures information destruction by adding noise tailored to the geometry of the loss landscape. We exploit the connections between the activation and weight dynamics of a DNN inspired by Neural Tangent Kernels to compute the information in the activations.
Has companion code repository: https://github.com/AdityaGolatkar/SelectiveForgetting
This page was built for publication: Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6336222)
