# End-to-End Learning of Geometrical Shaping Maximizing Generalized Mutual Information

arXiv: https://arxiv.org/abs/1912.05638 · HTML: https://mlsanity.com/paper/1912.05638/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Dec 2019.

## Built on by (most cited first, 4 of 4)

- [End-to-End Deep Learning of Long-Haul Coherent Optical Fiber Communications via Regular Perturbation Model](https://mlsanity.com/paper/2107.12320/) (2021)
- [End-to-End Learning of Joint Geometric and Probabilistic Constellation Shaping](https://mlsanity.com/paper/2112.05050/) (2021)
- [Geometric Constellation Shaping for Fiber-Optic Channels via End-to-End Learning](https://mlsanity.com/paper/2211.04311/) (2022)
- [Friendly Attacks to Improve Channel Coding Reliability](https://mlsanity.com/paper/2401.14184/) (2024)

## Abstract (arXiv)

GMI-based end-to-end learning is shown to be highly nonconvex. We apply gradient descent initialized with Gray-labeled APSK constellations directly to the constellation coordinates. State-of-the-art constellations in 2D and 4D are found providing reach increases up to 26% w.r.t. to QAM.