English

Improved lower bounds for the Shannon capacity of odd cycles

Information Theory 2026-07-23 v1 Artificial Intelligence Discrete Mathematics Combinatorics

Abstract

The Shannon capacity Θ(G)\Theta(G) of a graph GG quantifies the maximum rate at which information can be transmitted with zero error over a noisy channel. It is lower bounded by α(Gd)1/d\alpha(G^d)^{1/d} for any dd, where α(Gd)\alpha(G^d) is the independence number of the dd-th strong power of GG. We construct independent sets of size 134753134753 in C710C_7^{10}, 2190921909 in C116C_{11}^{6}, and 6253062530 in C136C_{13}^{6}, improving the best known lower bounds for the Shannon capacity of these graphs to Θ(C7)1347531/10>3.258020\Theta(C_7)\geq 134753^{1/10}>3.258020, Θ(C11)219091/6>5.289773\Theta(C_{11})\geq 21909^{1/6}>5.289773, and Θ(C13)625301/6>6.300109\Theta(C_{13})\geq 62530^{1/6}>6.300109. We also improve the best known lower bounds on the independence numbers of several individual strong powers of odd cycles that do not improve the Shannon capacity lower bound. The constructions were discovered through iterative interactions with a Large Language Model (LLM), illustrating the potential of LLMs for finding explicit combinatorial constructions.

Keywords

Cite

@article{arxiv.2607.21517,
  title  = {Improved lower bounds for the Shannon capacity of odd cycles},
  author = {Nathaniel Itty and Christopher D. Rosin and Chase Carstensen and Daniel Reichman},
  journal= {arXiv preprint arXiv:2607.21517},
  year   = {2026}
}