English

Generalization Bounds for Transformer Channel Decoders

Information Theory 2026-01-13 v1 Machine Learning math.IT

Abstract

Transformer channel decoders, such as the Error Correction Code Transformer (ECCT), have shown strong empirical performance in channel decoding, yet their generalization behavior remains theoretically unclear. This paper studies the generalization performance of ECCT from a learning-theoretic perspective. By establishing a connection between multiplicative noise estimation errors and bit-error-rate (BER), we derive an upper bound on the generalization gap via bit-wise Rademacher complexity. The resulting bound characterizes the dependence on code length, model parameters, and training set size, and applies to both single-layer and multi-layer ECCTs. We further show that parity-check-based masked attention induces sparsity that reduces the covering number, leading to a tighter generalization bound. To the best of our knowledge, this work provides the first theoretical generalization guarantees for this class of decoders.

Keywords

Cite

@article{arxiv.2601.06969,
  title  = {Generalization Bounds for Transformer Channel Decoders},
  author = {Qinshan Zhang and Bin Chen and Yong Jiang and Shu-Tao Xia},
  journal= {arXiv preprint arXiv:2601.06969},
  year   = {2026}
}

Comments

18 pages, 3 figures

R2 v1 2026-07-01T08:59:40.566Z