Search

Nowadays, recommender system is an indispensable tool in many information services, and a large number of algorithms have been designed and implemented. However, fed with very large datasets, state-of-the-art recommendation algorithms often face an efficiency bottleneck, i.e., it takes huge amount of computing resources to train a recommendation model. In order to satisfy the needs of privacy-savvy users who do not want to disclose their information to the service provider, the complexity of most existing solutions becomes prohibitive. As such, it is an interesting research question to design simple and efficient recommendation algorithms that achieve reasonable accuracy and facilitate privacy protection at the same time. In this paper, we propose an efficient recommendation algorithm, named CryptoRec, which has two nice properties: (1) can estimate a new user's preferences by directly using a model pre-learned from an expert dataset, and the new user's data is not required to train the model; (2) can compute recommendations with only addition and multiplication operations. As to the evaluation, we first test the recommendation accuracy on three real-world datasets and show that CryptoRec is competitive with state-of-the-art recommenders. Then, we evaluate the performance of the privacy-preserving variants of CryptoRec and show that predictions can be computed in seconds on a PC. In contrast, existing solutions will need tens or hundreds of hours on more powerful computers.

Machine-Learning-as-a-Service has become increasingly popular, with Recommendation-as-a-Service as one of the representative examples. In such services, providing privacy protection for the users is an important topic. Reviewing privacy-preserving solutions which were proposed in the past decade, privacy and machine learning are often seen as two competing goals at stake. Though improving cryptographic primitives (e.g., secure multi-party computation (SMC) or homomorphic encryption (HE)) or devising sophisticated secure protocols has made a remarkable achievement, but in conjunction with state-of-the-art recommender systems often yields far-from-practical solutions. We tackle this problem from the direction of machine learning. We aim to design crypto-friendly recommendation algorithms, thus to obtain efficient solutions by directly using existing cryptographic tools. In particular, we propose an HE-friendly recommender system, refer to as CryptoRec, which (1) decouples user features from latent feature space, avoiding training the recommendation model on encrypted data; (2) only relies on addition and multiplication operations, making the model straightforwardly compatible with HE schemes. The properties turn recommendation-computations into a simple matrix-multiplication operation. To further improve efficiency, we introduce a sparse-quantization-reuse method which reduces the recommendation-computation time by $9\times$ (compared to using CryptoRec directly), without compromising the accuracy. We demonstrate the efficiency and accuracy of CryptoRec on three real-world datasets. CryptoRec allows a server to estimate a user's preferences on thousands of items within a few seconds on a single PC, with the user's data homomorphically encrypted, while its prediction accuracy is still competitive with state-of-the-art recommender systems computing over clear data. Our solution enables Recommendation-as-a-Service on large datasets in a nearly real-time (seconds) level.

Functional encryption (FE) allows an authority to issue tokens associated with various functions, allowing the holder of some token for function f to learn only \(f(\mathsf {D})\) from a ciphertext that encrypts \(\mathsf {D}\). The standard approach is to model f as a circuit, which yields inefficient evaluations over large inputs. Here, we propose a new primitive that we call updatable functional encryption (UFE), where instead of circuits we deal with RAM programs, which are closer to how programs are expressed in von Neumann architecture. We impose strict efficiency constrains in that the run-time of a token \(\mathsf {\overline{P}}\) on ciphertext \(\mathsf {CT}\) is proportional to the run-time of its clear-form counterpart (program \(\mathsf {P}\) on memory \(\mathsf {D}\)) up to a polylogarithmic factor in the size of \(\mathsf {D}\), and we envision tokens that are capable to update the ciphertext, over which other tokens can be subsequently executed. We define a security notion for our primitive and propose a candidate construction from obfuscation, which serves as a starting point towards the realization of other schemes and contributes to the study on how to compute RAM programs over public-key encrypted data.

Lecture Notes in Computer Science, 7341, We extend the work of Bellare, Boldyreva and Staddon on the systematic analysis of randomness reuse to construct multi-recipient encryption schemes to the case where randomness is reused across different cryptographic primitives. We find that through the additional binding introduced through randomness reuse, one can actually obtain a security amplification with respect to the standard black-box compositions, and achieve a stronger level of security. We introduce stronger notions of security for encryption and signatures, where challenge messages can depend in a restricted way on the random coins used in encryption, and show that two variants of the KEM/DEM paradigm give rise to encryption schemes that meet this enhanced notion of security. We obtain the most efficient signcryption scheme to date that is secure against insider attackers without random oracles., (undefined)