Code automorphisms used for data augmentation during training and inference allow syndrome-based neural decoders to closely approach maximum likelihood performance on short high-rate codes with small datasets.
Efficient decoders for short block length codes in 6G URLLC
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.IT 2representative citing papers
The authors develop an error pattern tree representation that unifies SGRAND and ORBGRAND, enabling a parallel SGRAND design with ML optimality and an enhanced parallel ORBGRAND variant, with reported speedups of 3.96x and 4.21x respectively.
citing papers explorer
-
Leveraging Code Automorphisms for Improved Syndrome-Based Neural Decoding
Code automorphisms used for data augmentation during training and inference allow syndrome-based neural decoders to closely approach maximum likelihood performance on short high-rate codes with small datasets.
-
A Parallelization Strategy for GRAND with Optimality Guarantee by Exploiting Error Pattern Tree Representation
The authors develop an error pattern tree representation that unifies SGRAND and ORBGRAND, enabling a parallel SGRAND design with ML optimality and an enhanced parallel ORBGRAND variant, with reported speedups of 3.96x and 4.21x respectively.