Quantum homomorphic encryption (QHE) is an important branch of quantum cryptography. It can directly calculate the quantum ciphertext while ensuring calculation correctness and data security. QHE for a T-gate generates an additional S-gate. If this error is not eliminated, the desired output cannot be obtained. Using quantum gate teleportation can eliminate the S-error non-interactively, but increases the decryption complexity. This paper uses the rotation operator to realize QHE for the T / T^† -gate, and proposes a non-interactive QHE scheme. The decryption complexity of this scheme is O(1), and the encryption complexity is O(N) , where N is the number of quantum gates in the evaluated circuit. We prove that the scheme is information-theoretic secure and mathcalF-homomorphic, i.e., homomorphic for any quantum circuit, and implement QHE for a Toffoli-gate decomposition circuit on an IBM Quantum Experience platform.
Quantum machine learning in cloud environments requires protecting sensitive data while enabling remote computation. Here we demonstrate the first realistic implementations of a perfectly-secure quantum homomorphic encryption (QHE) scheme applied to quantum neural networks (QNN). Using efficient Clifford+
T decomposition, we implement quantum convolutional neural networks for two complementary scenarios: (i) reverse delegated training, where encrypted data from multiple providers trains a user's network via federated aggregation; (ii) private inference, where users process encrypted data with remote quantum networks. Moreover, analysis of server circuit privacy reveals probabilistic model protection through Pauli gate concealment. These results establish perfectly-secure QHE as a practical framework for multi-party quantum machine learning.
Suppose some data have been encrypted, can you compute with the data without decrypting them? This problem has been studied as homomorphic encryption and blind computing. We consider this problem in the context of quantum information processing, and present the definitions of quantum homomorphic encryption (QHE) and quantum fully homomorphic encryption (QFHE). Then, based on quantum one-time pad (QOTP), we construct a symmetric QFHE scheme, where the evaluate algorithm depends on the secret key. This scheme permits any unitary transformation on any
n-qubit state that has been encrypted. Compared with classical homomorphic encryption, the QFHE scheme has perfect security. Finally, we also construct a QOTP-based symmetric QHE scheme, where the evaluate algorithm is independent of the secret key.