UC San Diego

In list-decodable learning, we are given a set of data points such that an α-fraction of these points come from a “nice” distribution D, for some small α ≪ 1, and the goal is to output a short list of candidate solutions, such that at least one element of this list recovers some non-trivial information about D. By now, there is a large body of work on this topic; however, while many algorithms can achieve optimal list size in terms of α, all known algorithms must incur error which decays, in some cases quite poorly, with 1 / α. In this paper, we ask if this is inherent: is it possible to trade off list size with accuracy in list-decodable learning? More formally, given ε > 0, can we output a slightly larger list in terms of α and ε, but so that one element of this list has error at most ε with the ground truth? We call this problem high-accuracy list-decodable learning. Our main result is that non-trivial high-accuracy guarantees, both information-theoretically and algorithmically, are possible for the canonical setting of list-decodable mean estimation of identity-covariance Gaussians. Specifically, we demonstrate that there exists a list of candidate means of size at most L = exp( O( log2 1 / α/ε2 )) so that one of the elements of this list has ℓ2 distance at most ε to the true mean. We also design an algorithm that outputs such a list with runtime and sample complexity n = dO(logL) + expexp(O(logL)). In particular, our results demonstrate that in the natural regime where α and ε are both small constants, it is possible to achieve error ≤ 0.01 in fully-polynomial time, where all prior work suffered error which was much larger than 1. We do so by demonstrating a completely novel proof of identifiability, as well as a new algorithmic way of leveraging this proof without the sum-of-squares hierarchy, which may be of independent technical interest.