English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Cloud providers are extending support for trusted hardware primitives such asIntel SGX. Simultaneously, the field of deep learning is seeing enormousinnovation as well as an increase in adoption. In this paper, we ask a timelyquestion: "Can third-party cloud services use Intel SGX enclaves to providepractical, yet secure DNN Inference-as-a-service?" We first demonstrate thatDNN models executing inside enclaves are vulnerable to access pattern basedattacks. We show that by simply observing access patterns, an attacker canclassify encrypted inputs with 97% and 71% attack accuracy for MNIST andCIFAR10 datasets on models trained to achieve 99% and 79% original accuracyrespectively. This motivates the need for PRIVADO, a system we have designedfor secure, easy-to-use, and performance efficient inference-as-a-service.PRIVADO is input-oblivious: it transforms any deep learning framework that iswritten in C/C++ to be free of input-dependent access patterns thus eliminatingthe leakage. PRIVADO is fully-automated and has a low TCB: with zero developereffort, given an ONNX description of a model, it generates compact andenclave-compatible code which can be deployed on an SGX cloud platform. PRIVADOincurs low performance overhead: we use PRIVADO with Torch framework and showits overhead to be 17.18% on average on 11 different contemporary neuralnetworks.

Privado: Practical and Secure DNN Inference with Enclaves