Machine unlearning has raised significant interest with the adoption of laws ensuring the "right to be forgotten".

"The right to be forgotten" ensured by laws for user data privacy becomes increasingly important.

In this work, we establish rapid convergence for these algorithms under distance measures more suitable for differential privacy.

Sample-dependent priors are also relevant in emerging scenarios such as adversarial training. A common practice here is to smooth a given classifier by injecting Gaussian noise into the inputs.

A soft-max function is a mechanism for choosing one out of a number of options, given the value of each option.