Sigma: Secure gpt inference with function secret sharing
K Gupta, N Jawalkar, A Mukherjee… - Cryptology ePrint …, 2023 - eprint.iacr.org
Abstract Secure 2-party computation (2PC) enables secure inference that offers protection
for both proprietary machine learning (ML) models and sensitive inputs to them. However …
for both proprietary machine learning (ML) models and sensitive inputs to them. However …
Grotto: Screaming fast (2+ 1)-PC or ℤ2n via (2, 2)-DPFs
We introduce Grotto, a framework and C++ library for space-and time-efficient (2+ 1)-party
piecewise polynomial (ie, spline) evaluation on secrets additively shared over ℤ2n. Grotto …
piecewise polynomial (ie, spline) evaluation on secrets additively shared over ℤ2n. Grotto …
Half-tree: Halving the cost of tree expansion in cot and dpf
GGM tree is widely used in the design of correlated oblivious transfer (COT), subfield vector
oblivious linear evaluation (sVOLE), distributed point function (DPF), and distributed …
oblivious linear evaluation (sVOLE), distributed point function (DPF), and distributed …
Secure softmax/sigmoid for machine-learning computation
Softmax and sigmoid, composing exponential functions (ex) and division (1/x), are activation
functions often required in training. Secure computation on non-linear, unbounded 1/x and …
functions often required in training. Secure computation on non-linear, unbounded 1/x and …
Falcon: Accelerating homomorphically encrypted convolutions for efficient private mobile network inference
Efficient networks, eg, MobileNetV2, EfficientNet, etc, achieves state-of-the-art (SOTA)
accuracy with lightweight computation. However, existing homomorphic encryption (HE) …
accuracy with lightweight computation. However, existing homomorphic encryption (HE) …
Orca: FSS-based Secure Training and Inference with GPUs
Abstract Secure Two-party Computation (2PC) allows two parties to compute any function on
their private inputs without revealing their inputs to each other. In the offline/online model for …
their private inputs without revealing their inputs to each other. In the offline/online model for …
CompactTag: Minimizing Computation Overheads in Actively-Secure MPC for Deep Neural Networks
Secure Multiparty Computation (MPC) protocols enable secure evaluation of a circuit by
several parties, even in the presence of an adversary who maliciously corrupts all but one of …
several parties, even in the presence of an adversary who maliciously corrupts all but one of …
Communication-efficient secure logistic regression
We present a new construction for secure logistic regression training, which enables two
parties to train a model on private secret-shared data. Our goal is to minimize online …
parties to train a model on private secret-shared data. Our goal is to minimize online …
HEQuant: Marrying Homomorphic Encryption and Quantization for Communication-Efficient Private Inference
Secure two-party computation with homomorphic encryption (HE) protects data privacy with
a formal security guarantee but suffers from high communication overhead. While previous …
a formal security guarantee but suffers from high communication overhead. While previous …
Tabula: Efficiently computing nonlinear activation functions for secure neural network inference
Multiparty computation approaches to secure neural network inference traditionally rely on
garbled circuits for securely executing nonlinear activation functions. However, garbled …
garbled circuits for securely executing nonlinear activation functions. However, garbled …