Encrypted CNN

Encrypted CNN refers to the use of cryptographic techniques to ensure the privacy of the data while performing convolutional neural network (CNN) computations. This approach allows two or more parties to jointly compute a CNN model over their private data without revealing the data to each other. One common technique used for encrypted CNN is homomorphic encryption, which allows computation on encrypted data without decrypting it first. Encrypted CNN has potential applications in secure image recognition, medical data analysis, and other fields where data privacy is a concern.