A field of research exploring how to apply zero-knowledge proofs to machine learning models, potentially enabling privacy-preserving training and inference.
