Description
The "harvest-now, decrypt-later" threat has spurred the search for post-quantum cryptographic schemes whose security should hold up against quantum attack. Of the candidates, lattice-based constructions have emerged as the leading family, built on problems such as the Shortest Vector Problem and Learning With Errors (LWE). These constructions provide a foundation for post-quantum security, since no efficient quantum algorithm is yet known that could break their presumed-hard problems.
Homomorphic computation, meanwhile, broadens privacy protection from static data to operations carried out directly on ciphertext. Especially notable is its pairing with privacy-preserving machine learning, where encrypted computation and techniques like differential privacy work together to curb the leakage of information from underlying training datasets.
| I am the presenting author | Yes |
|---|