Improved List-Decodability of Reed--Solomon Codes via Tree Packings

Abstract: This paper shows that there exist Reed--Solomon (RS) codes, over \black{exponentially} large finite fields \black{in the code length}, that are combinatorially list-decodable well beyond the Johnson radius, in fact almost achieving the list-decoding capacity. In particular, we show that for any $\epsilon\in (0,1]$ there exist RS codes with rate $\Omega(\frac{\epsilon}{\log(1/\epsilon)+1})$ that are list-decodable from radius of $1-\epsilon$. We generalize this result to list-recovery, showing that there exist $(1 - \epsilon, \ell, O(\ell/\epsilon))$-list-recoverable RS codes with rate $$\Omega\left( \frac{\epsilon}{\sqrt{\ell} (\log(1/\epsilon)+1)} \right)$$. Along the way we use our techniques to give a new proof of a result of Blackburn on optimal linear perfect hash matrices, and strengthen it to obtain a construction of strongly perfect hash matrices. To derive the results in this paper we show a surprising connection of the above problems to graph theory, and in particular to the tree packing theorem of Nash-Williams and Tutte. We also state a new conjecture that generalizes the tree-packing theorem to hypergraphs, and show that if this conjecture holds, then there would exist RS codes that are \em optimally \em (non-asymptotically) list-decodable.