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

ML-as-a-service is gaining popularity where a cloud server hosts a trainedmodel and offers prediction (inference) service to users. In this setting, ourobjective is to protect the confidentiality of both the users' input queries aswell as the model parameters at the server, with modest computation andcommunication overhead. Prior solutions primarily propose fine-tuningcryptographic methods to make them efficient for known fixed modelarchitectures. The drawback with this line of approach is that the model itselfis never designed to operate with existing efficient cryptographiccomputations. We observe that the network architecture, internal functions, andparameters of a model, which are all chosen during training, significantlyinfluence the computation and communication overhead of a cryptographic method,during inference. Based on this observation, we propose SOTERIA -- a trainingmethod to construct model architectures that are by-design efficient forprivate inference. We use neural architecture search algorithms with the dualobjective of optimizing the accuracy of the model and the overhead of usingcryptographic primitives for secure inference. Given the flexibility ofmodifying a model during training, we find accurate models that are alsoefficient for private computation. We select garbled circuits as our underlyingcryptographic primitive, due to their expressiveness and efficiency, but thisapproach can be extended to hybrid multi-party computation settings. Weempirically evaluate SOTERIA on MNIST and CIFAR10 datasets, to compare with theprior work. Our results confirm that SOTERIA is indeed effective in balancingperformance and accuracy.

SOTERIA: In Search of Efficient Neural Networks for Private Inference