Crypto-oriented neural architecture design

Avital Shafran, Gil Segev, Shmuel Peleg, Yedid Hoshen

Research output: Contribution to journalConference articlepeer-review


Sending private data to Neural Network applications raises many privacy concerns. The cryptography community developed a variety of secure computation methods to address such privacy issues. As generic techniques for secure computation are typically prohibitively expensive, efforts focus on optimizing these cryptographic tools. Differently, we propose to optimize the design of crypto-oriented neural architectures, introducing a novel Partial Activation layer. The proposed layer is much faster for secure computation as it contains fewer non linear computations. Evaluating our method on three state-of-the-art architectures (SqueezeNet, ShuffleNetV2, and MobileNetV2) demonstrates significant improvement to the efficiency of secure inference on common evaluation metrics.

Original languageAmerican English
Pages (from-to)2680-2684
Number of pages5
JournalProceedings - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing
StatePublished - 2021
Event2021 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2021 - Virtual, Toronto, Canada
Duration: 6 Jun 202111 Jun 2021


  • NN Architecture
  • Secure Inference

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Electrical and Electronic Engineering


Dive into the research topics of 'Crypto-oriented neural architecture design'. Together they form a unique fingerprint.

Cite this