On subexponential running times for approximating directed Steiner tree and related problems

by   Marek Cygan, et al.

This paper concerns proving almost tight (super-polynomial) running times, for achieving desired approximation ratios for various problems. To illustrate, the question we study, let us consider the Set-Cover problem with n elements and m sets. Now we specify our goal to approximate Set-Cover to a factor of (1-d)ln n, for a given parameter 0<d<1. What is the best possible running time for achieving such approximation? This question was answered implicitly in the work of Moshkovitz [Theory of Computing, 2015]: Assuming both the Projection Games Conjecture (PGC) and the Exponential-Time Hypothesis (ETH), any ((1-d) ln n)-approximation algorithm for Set-Cover must run in time >= 2^n^c d, for some constant 0<d<1. We study the questions along this line. First, we show that under ETH and PGC any ((1-d) n)-approximation for Set-Cover requires 2^n^d-time. This (almost) matches the running time of 2^O(n^d) for approximating Set-Cover to a factor (1-d) ln n by Cygan et al. [IPL, 2009]. Our result is tight up to the constant multiplying the n^d terms in the exponent. This lower bound applies to all of its generalizations, e.g., Group Steiner Tree (GST), Directed Steiner (DST), Covering Steiner Tree (CST), Connected Polymatroid (CP). We also show that in almost exponential time, these problems reduce to Set-Cover: We show (1-d)ln n approximation algorithms for all these problems that run in time 2^n^d n poly(m). We also study log^2-dn approximation for GST. Chekuri-Pal [FOCS, 2005] showed that GST admits (log^2-dn)-approximation in time exp(2^log^d+o(1)n), for any 0 < d < 1. We show the lower bound of GST: any (log^2-dn)-approximation for GST must run in time >= exp((1+o(1))log^d-cn), for any c>0, unless the ETH is false. Our result follows by analyzing the work of Halperin and Krauthgamer [STOC, 2003]. The same lower and upper bounds hold for CST.


page 1

page 2

page 3

page 4


Tight Bounds on Subexponential Time Approximation of Set Cover and Related Problems

We show that Set Cover on instances with N elements cannot be approximat...

Nearly Optimal Time Bounds for kPath in Hypergraphs

In the kPath problem the input is a graph G and an integer k, and the go...

Approximating Sparsest Cut in Low-Treewidth Graphs via Combinatorial Diameter

The fundamental sparsest cut problem takes as input a graph G together w...

Covering convex bodies and the Closest Vector Problem

We present algorithms for the (1+ϵ)-approximate version of the closest v...

Near-Optimal Algorithms for Point-Line Covering Problems

We study fundamental point-line covering problems in computational geome...

Revisiting the Set Cover Conjecture

In the Set Cover problem, the input is a ground set of n elements and a ...

Optimal-Degree Polynomial Approximations for Exponentials and Gaussian Kernel Density Estimation

For any real numbers B ≥ 1 and δ∈ (0, 1) and function f: [0, B] →ℝ, let ...

Please sign up or login with your details

Forgot password? Click here to reset