New Sublinear Algorithms and Lower Bounds for LIS Estimation

TL;DR

This paper establishes new lower bounds and algorithms for estimating the LIS in arrays, revealing fundamental limits and providing more efficient methods especially when array values are limited.

Contribution

It presents the first nontrivial lower bounds for LIS estimation and introduces novel nonadaptive algorithms that adapt to the number of distinct values and LIS length.

Findings

Nonadaptive algorithms require polylogarithmic queries for additive error bounds.

Algorithms' complexity decreases with fewer distinct array values.

A new erasure-resilient sortedness tester with logarithmic queries.

Abstract

Estimating the length of the longest increasing subsequence (LIS) in an array is a problem of fundamental importance. Despite the significance of the LIS estimation problem and the amount of attention it has received, there are important aspects of the problem that are not yet fully understood. There are no better lower bounds for LIS estimation than the obvious bounds implied by testing monotonicity (for adaptive or nonadaptive algorithms). In this paper, we give the first nontrivial lower bound on the complexity of LIS estimation, and also provide novel algorithms that complement our lower bound. Specifically, for every constant $\epsilon \in (0,1)$, every nonadaptive algorithm that outputs an estimate of the length of the LIS in an array of length $n$ to within an additive error of $\epsilon \cdot n$ has to make $\log^{\Omega(\log (1/\epsilon))} n)$ queries. Next, we design nonadaptive LIS estimation algorithms whose complexity decreases as the the number of distinct values, $r$, in the array decreases. We first present a simple algorithm that makes $\tilde{O}(r/\epsilon^3)$ queries and approximates the LIS length with an additive error bounded by $\epsilon n$. We then use it to construct a nonadaptive algorithm with query complexity $\tilde{O}(\sqrt{r} \cdot \text{poly}(1/\lambda))$ that, for an array with LIS length at least $\lambda n$, outputs a multiplicative $\Omega(\lambda)$-approximation to the LIS length. Finally, we describe a nonadaptive erasure-resilient tester for sortedness, with query complexity $O(\log n)$. Our result implies that nonadaptive tolerant testing is strictly harder than nonadaptive erasure-resilient testing for the natural property of monotonicity.