Security
Protection
Techniques that protect a model or its data from attack, theft or exposure, such as encryption or ownership marking.
3 techniques
3 techniques
| Goals | Models | Data Types | Description | |||
|---|---|---|---|---|---|---|
| Homomorphic Encryption | Algorithmic | Architecture/linear Models Architecture/model Agnostic +5 | Any | Homomorphic encryption lets a server compute on data that stays encrypted, so the server never sees the underlying... | ||
| Trigger-Set Model Watermarking | Algorithmic | Architecture/neural Networks Paradigm/supervised +1 | Image | Trigger-set model watermarking lets the owner of a trained classifier show later that a suspect model was copied from... | ||
| Model Signing | Algorithmic | Architecture/model Agnostic Requirements/white Box | Any | Model signing lets the publisher of a machine-learning model attach a cryptographic signature to the model's files, so... |
Rows per page
Page 1 of 1