Learning-augmented Maximum Independent Set

Braverman, Vladimir, Dharangutte, Prathamesh, Shah, Vihan, Wang, Chen

arXiv.org Artificial Intelligence 

We study the Maximum Independent Set (MIS) problem on general graphs within the framework of learning-augmented algorithms. The MIS problem is known to be NP-hard and is also NP-hard to approximate to within a factor of $n^{1-\delta}$ for any $\delta>0$. We show that we can break this barrier in the presence of an oracle obtained through predictions from a machine learning model that answers vertex membership queries for a fixed MIS with probability $1/2+\varepsilon$. In the first setting we consider, the oracle can be queried once per vertex to know if a vertex belongs to a fixed MIS, and the oracle returns the correct answer with probability $1/2 + \varepsilon$. Under this setting, we show an algorithm that obtains an $\tilde{O}(\sqrt{\Delta}/\varepsilon)$-approximation in $O(m)$ time where $\Delta$ is the maximum degree of the graph. In the second setting, we allow multiple queries to the oracle for a vertex, each of which is correct with probability $1/2 + \varepsilon$. For this setting, we show an $O(1)$-approximation algorithm using $O(n/\varepsilon^2)$ total queries and $\tilde{O}(m)$ runtime.

Duplicate Docs Excel Report

Title
None found

Similar Docs  Excel Report  more

TitleSimilaritySource
None found