Joint End-to-End Training of Weighted BP Decoder Ensembles
Abstract
Ensemble decoding improves decoding reliability by combining component decoders with complementary failure events. Existing approaches typically obtain this diversity from code-specific structure, such as code automorphisms, multiple parity-check representations, or targeted modifications of the message-passing graph. In this work, we instead learn the diversity of an ensemble of weighted belief propagation decoders by jointly optimizing trainable Tanner-graph edge weights. A central design question is how the training objective should distribute gradient among the ensemble members to promote complementary decoding behavior. To this end, we compare four ensemble objectives under direct gradient-based optimization and identify best-member training as the most effective approach. For a (64,32) progressive edge-growth (PEG) low-density parity-check (LDPC) code, the resulting ensemble outperforms the considered trained and untrained belief-propagation-based ensemble baselines. With 10 members, it provides a gain of approximately 0.66 dB over a single belief propagation decoder at a target frame error rate (FER) of 10−4. Similar behavior is observed for a (128,64) CCSDS LDPC code, indicating that the learned complementarity is not specific to PEG. Although the ensemble performance improves substantially, all trained members show a higher standalone FER. The gain therefore results from learned specialization and complementary failure events rather than uniformly improved component decoders.
Related
- Random Matrix Spectra from Boltzmann-Weighted Lattice Ensembles
- From Pixels, Without Pre-training: Joint Generative and Self-Supervised Representation Learning in One Model
- Pathfinder-BP: A Low-Latency Recursive Decoder for Topological Codes
- Four-Level Quantized Soft Information Aided Hard-Decision Decoder for Product-like Codes
- Early Selection of Polar Ensemble Decoding with Successive Cancellation Style Component Decoder
BibTeX
@misc{krieg2026joint,
title = {{Joint End-to-End Training of Weighted BP Decoder Ensembles}},
author = {Felix Krieg and Paul Bezner and Stephan ten Brink},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.33559},
eprint = {2610.33559},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.33559},
}
RIS
TY - UNPB
TI - Joint End-to-End Training of Weighted BP Decoder Ensembles
AU - Felix Krieg
AU - Paul Bezner
AU - Stephan ten Brink
PY - 2026
DA - 2026/10/24
PB - arXiv
DO - 10.48550/arxiv.2610.33559
UR - https://arxiv.org/abs/2610.33559
AB - Ensemble decoding improves decoding reliability by combining component decoders with complementary failure events. Existing approaches typically obtain this diversity from code-specific structure, such as code automorphisms, multiple parity-check representations, or targeted modifications of the message-passing graph. In this work, we instead learn the diversity of an ensemble of weighted belief propagation decoders by jointly optimizing trainable Tanner-graph edge weights. A central design question is how the training objective should distribute gradient among the ensemble members to promote complementary decoding behavior. To this end, we compare four ensemble objectives under direct gradient-based optimization and identify best-member training as the most effective approach. For a (64,32) progressive edge-growth (PEG) low-density parity-check (LDPC) code, the resulting ensemble outperforms the considered trained and untrained belief-propagation-based ensemble baselines. With 10 members, it provides a gain of approximately 0.66 dB over a single belief propagation decoder at a target frame error rate (FER) of 10^−4. Similar behavior is observed for a (128,64) CCSDS LDPC code, indicating that the learned complementarity is not specific to PEG. Although the ensemble performance improves substantially, all trained members show a higher standalone FER. The gain therefore results from learned specialization and complementary failure events rather than uniformly improved component decoders.
ER -