Saved in:
Bibliographic Details
Main Authors: Srivastava, Shubham, Banerjee, Adrish
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2506.10121
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918055628505088
author Srivastava, Shubham
Banerjee, Adrish
author_facet Srivastava, Shubham
Banerjee, Adrish
contents This paper introduces HiKO (Hierarchical Kronecker Operation), a novel framework for training high-rate neural error-correcting codes that enables KO codes to outperform Reed-Muller codes beyond second order. To our knowledge, this is the first attempt to extend KO codes beyond second order. While conventional KO codes show promising results for low-rate regimes ($r < 2$), they degrade at higher rates -- a critical limitation for practical deployment. Our framework incorporates three key innovations: (1) a hierarchical training methodology that decomposes complex high-rate codes into simpler constituent codes for efficient knowledge transfer, (2) enhanced neural architectures with dropout regularization and learnable skip connections tailored for the Plotkin structure, and (3) a progressive unfreezing strategy that systematically transitions from pre-trained components to fully optimized integrated codes. Our experiments show that HiKO codes consistently outperform traditional Reed-Muller codes across various configurations, achieving notable performance improvements for third-order ($r = 3$) and fourth-order ($r = 4$) codes. Analysis reveals that HiKO codes successfully approximate Shannon-optimal Gaussian codebooks while preserving efficient decoding properties. This represents the first successful extension of KO codes beyond second order, opening new possibilities for neural code deployment in high-throughput communication systems.
format Preprint
id arxiv_https___arxiv_org_abs_2506_10121
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HiKO: A Hierarchical Framework for Beyond-Second-Order KO Codes
Srivastava, Shubham
Banerjee, Adrish
Information Theory
This paper introduces HiKO (Hierarchical Kronecker Operation), a novel framework for training high-rate neural error-correcting codes that enables KO codes to outperform Reed-Muller codes beyond second order. To our knowledge, this is the first attempt to extend KO codes beyond second order. While conventional KO codes show promising results for low-rate regimes ($r < 2$), they degrade at higher rates -- a critical limitation for practical deployment. Our framework incorporates three key innovations: (1) a hierarchical training methodology that decomposes complex high-rate codes into simpler constituent codes for efficient knowledge transfer, (2) enhanced neural architectures with dropout regularization and learnable skip connections tailored for the Plotkin structure, and (3) a progressive unfreezing strategy that systematically transitions from pre-trained components to fully optimized integrated codes. Our experiments show that HiKO codes consistently outperform traditional Reed-Muller codes across various configurations, achieving notable performance improvements for third-order ($r = 3$) and fourth-order ($r = 4$) codes. Analysis reveals that HiKO codes successfully approximate Shannon-optimal Gaussian codebooks while preserving efficient decoding properties. This represents the first successful extension of KO codes beyond second order, opening new possibilities for neural code deployment in high-throughput communication systems.
title HiKO: A Hierarchical Framework for Beyond-Second-Order KO Codes
topic Information Theory
url https://arxiv.org/abs/2506.10121