Privacy
8 concepts, 8 reviewed and merged.
Attacks
- Gradient LeakageFederated learning sends gradients instead of data on the assumption that a gradient reveals little.verified
- Membership InferenceGiven a trained model and a record, decide whether that record was in the training set.verified
- Training Data ExtractionA language model can be made to emit sequences it memorised verbatim — addresses, keys, phone numbers — by prompting…verified