CWI Cryptology Group

Abstract: A recommendation system involves a large number of users that each initially provide a list of ratings or preferences. During the lifetime of the system users are both able to update their provided list and to request an up-to-date personalized list of recommendations based on the most similar users in the system. In this talk I will present some technical details of ongoing work on secure recommendation systems, where none of the data provided by the users should leak beyond what is exposed by the provided recommendations.

Abstract: Physical Unclonable Functions (PUFs) are an emerging, alternative approach in cryptography and security. While their initial use was mainly as a novel key storage element, recent work has discovered their potential as a general cryptographic primitive, comparable to the bounded storage model, quantum cryptography, or noise-based cryptography. In this talk, we undertake a journey through the area, describing PUF-based protocols for identification, key exchange and oblivious transfer. We then move on to two new concepts, which improve and generalize PUFs in certain aspects: So-called SIMPL systems, which are a public key variant of PUFs; and so-termed Virtual Proofs of Reality (VPs), which aim at proving physical statements via remote communication. For future hardware implementations of these two concepts, quantum technology could be a very interesting candidate.

Abstract: This presentation considers encrypted computation where the user specifies encrypted inputs to an untrusted program, and the server computes on those encrypted inputs. To this end we propose a secure processor architecture, called Ascend, that guarantees privacy of data when arbitrary programs use the data running in a cloud-like environment (e.g., an untrusted server running an untrusted software stack).
The key idea to guarantee privacy is obfuscated instruction execution; Ascend does not disclose what instruction is being run at any given time, be it an arithmetic instruction or a memory instruction. Periodic accesses to external instruction and data memory are performed through an Oblivious RAM (ORAM) interface to prevent leakage through memory access patterns.
We evaluate the processor architecture on SPEC benchmarks running on encrypted data and quantify overheads.