search

kdmalc / personalization-privacy-risk

Privacy analysis for ML and classical filtering personalization parameters
0 stars 0 forks source link

Get real hessian version of APFL to run #32

Closed kdmalc closed 1 year ago

kdmalc commented 1 year ago

Fixed by commit 31507cce37d48394bd935ca95e0dfc1758fae074

kdmalc commented 1 year ago

Note that it runs but I'm not sure about real-time (had to add the hack about not recomputing the Hessian each time). Also note that the gradient is not quadratic... assuming this is a personalization/mixing issue. Real hess version runs, any future dev for APFL will be commented on #2