mirror of
https://github.com/zoo-apps/papers.git
synced 2026-11-02 17:38:34 +00:00
zeekay
1474380e17
Locked canonical naming for the Lux FHE product line: Lux FHE = FHE framework umbrella Torus = compiler / runtime TorusVM = core FHE runtime virtual machine TorusEVM = encrypted EVM (this rename) TorusML = private ML TorusNet = threshold/decryption network Zoo FHE / zoo-fhe-ai paper now references TorusEVM as the on-chain FHE-EVM that Zoo extends with CKKS support + ML inference at the T-Chain layer.
12 lines
958 B
TeX
12 lines
958 B
TeX
\section{Related Work}
|
|
|
|
|
|
\textbf{FHE for ML}: CryptoNets~\cite{gilad2016cryptonets} demonstrated CNN inference on encrypted data using leveled HE. Subsequent work improved performance~\cite{brutzkus2019low} but remained off-chain.
|
|
|
|
\textbf{On-chain FHE}: TorusEVM~\cite{lux-fhevm} brings \TFHE{} to the EVM via native precompiles at \texttt{0x...80}--\texttt{0x...83} with consensus-deterministic keygen. Zoo FHE extends this with \CKKS{} support for efficient linear algebra (ML inference) and threshold decryption at the \TChain{} layer.
|
|
|
|
\textbf{Federated Learning}: McMahan et al.~\cite{mcmahan2017communication} introduced FedAvg for distributed training. Zoo FHE adds encryption to prevent gradient leakage attacks~\cite{zhu2019deep}.
|
|
|
|
\textbf{Conservation AI}: The Zoo Network's federated wildlife monitoring system~\cite{zoo-federated-wildlife} and conservation AI framework~\cite{zoo-conservation-ai} provide the application context for this work.
|
|
|