Sequential Neural Probabilistic Amplitude Shaping: Learning the Channel's Language
Machine Learning
2026-05-28 v1 Information Theory
Signal Processing
math.IT
Abstract
We present the first neural probabilistic amplitude shaping that outperforms existing methods while accounting for all implementation losses, using a block-less, easily implementable sequential autoregressive encoder compatible with arithmetic distribution matching, yielding reduced rate loss and higher achievable information rates.
Cite
@article{arxiv.2605.28143,
title = {Sequential Neural Probabilistic Amplitude Shaping: Learning the Channel's Language},
author = {Mohammad Taha Askari and Lutz Lampe and Amirhossein Ghazisaeidi},
journal= {arXiv preprint arXiv:2605.28143},
year = {2026}
}
Comments
4 pages, 2 figures, Submitted to the 52nd European Conference on Optical Communications