Scaling Bayesian Bandit Encoding with Shared Learning
Abstract
A communication system must choose error protection and decoding effort as channel conditions change. A Bayesian bandit encoder (BBE) uses receiver feedback to learn which transmission configuration to select. We study a receiver that decodes by guessing error patterns, using Guessing Random Additive Noise Decoding (GRAND). We extend BBE's selection component to 1,008 code and decoder configurations by learning shared performance patterns offline and updating their weights online. Decoder noise models remain fixed. On a six-configuration training-selected shortlist, sharing reduces accumulated utility loss by 33.5% relative to independent learning. A fixed training-selected configuration matches the shared learner that searches the full catalog. After channel changes, the pruned shared learner first meets a near-optimal selection criterion in 88.5% of events by 2,000 packets, compared with 54.2% for pruned independent learning with the same discounting. The results support combining sharing and pruning for configuration selection, although packet losses remain high for the tested codes under severe noise.
Related
- Efficient Cost-Aware LLM Evaluation via Bayesian Bandit Gittins Indices
- Decentralized Bandit Learning of Markov Perfect Equilibria
- Detection Augmented Learning In Non-Stationary Bandit Environments
- PAC Learning with Bandit Feedback: Sharp Sample Complexity in the Realizable Setting
- Bandit Learning in General Open Multi-agent Systems
BibTeX
@misc{krishnamachari2026scaling,
title = {{Scaling Bayesian Bandit Encoding with Shared Learning}},
author = {Bhaskar Krishnamachari},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.06293},
eprint = {2609.06293},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.06293},
}
RIS
TY - UNPB
TI - Scaling Bayesian Bandit Encoding with Shared Learning
AU - Bhaskar Krishnamachari
PY - 2026
DA - 2026/09/05
PB - arXiv
DO - 10.48550/arxiv.2609.06293
UR - https://arxiv.org/abs/2609.06293
AB - A communication system must choose error protection and decoding effort as channel conditions change. A Bayesian bandit encoder (BBE) uses receiver feedback to learn which transmission configuration to select. We study a receiver that decodes by guessing error patterns, using Guessing Random Additive Noise Decoding (GRAND). We extend BBE's selection component to 1,008 code and decoder configurations by learning shared performance patterns offline and updating their weights online. Decoder noise models remain fixed. On a six-configuration training-selected shortlist, sharing reduces accumulated utility loss by 33.5% relative to independent learning. A fixed training-selected configuration matches the shared learner that searches the full catalog. After channel changes, the pruned shared learner first meets a near-optimal selection criterion in 88.5% of events by 2,000 packets, compared with 54.2% for pruned independent learning with the same discounting. The results support combining sharing and pruning for configuration selection, although packet losses remain high for the tested codes under severe noise.
ER -