Exact Unlearning via Quantized Sufficient Statistics
Exact unlearning requires a deployed predictor to match one rebuilt without the information named by a deletion request. Existing general-purpose exact methods localize retraining through disjoint shards, but every request still invalidates a model, and smaller shards reduce the data available to each constituent predi...
A. Tavory, Shripad Gade, Tal Sarig et al.
· 0 citations