Skip to Main content Skip to Navigation
Conference papers

An Turyn-based neural Leech decoder

Abstract : A new decoder for the Leech lattice is presented. This quasi-optimal decoder utilizes a re-encoding paradigm, where candidates are obtained via a shallow neural network. This implies easy parallelization and low latency. The decoder exploits the fact that the Leech lattice is obtained from the direct sum of three polarized Gosset 8-dimensional lattices. This Turyn's construction was used in 2010 by G. Nebe to build the extremal even unimodular lattice in dimension 72 from three copies of the Leech lattice. Thus, we view this work as a first step towards the implementation of an efficient decoder for the Nebe 72-dimensional lattice.
Complete list of metadata

Cited literature [28 references]  Display  Hide  Download
Contributor : Philippe CIBLAT Connect in order to contact the contributor
Submitted on : Monday, August 17, 2020 - 12:45:54 PM
Last modification on : Wednesday, November 3, 2021 - 6:22:44 AM
Long-term archiving on: : Tuesday, December 1, 2020 - 6:42:55 PM


Files produced by the author(s)


  • HAL Id : hal-02916090, version 1


Vincent Corlay, Joseph Boutros, Philippe Ciblat, Loïc Brunel. An Turyn-based neural Leech decoder. International Workshop on Coding and Cryptography (WCC), 2019, Saint-Jacut-de-la-Mer, France. ⟨hal-02916090⟩



Record views


Files downloads