Simple dynamic algorithms for Maximal Independent Set and other problems

Abstract

Most graphs in real life keep changing with time. These changes can be in the form of insertion or deletion of edges or vertices. Such rapidly changing graphs motivate us to study dynamic graph algorithms. However, three important graph problems that are perhaps not sufficiently addressed in the literature include independent sets, maximum matching (exact) and maximum flows. Maximal Independent Set (MIS) is one of the most prominently studied problems in the distributed setting. Recently, the first dynamic MIS algorithm for distributed networks was given by Censor-Hillel et al. [PODC16], requiring expected $O(1)$ amortized rounds with $O(\Delta)$ messages per update, where $\Delta$ is the maximum degree of a vertex in the graph. They suggested an open problem to maintain MIS in fully dynamic centralized setting more efficiently. Assadi et al. [STOC18] presented a deterministic centralized fully dynamic MIS algorithm requiring $O(\min\{\Delta,m^{3/4}\})$ amortized time per update. This result is quite complex involving an exhaustive case analysis. We report a surprisingly simple deterministic centralized algorithm which improves the amortized update time to $O(\min\{\Delta,m^{2/3}\})$. Additionally, we present some other minor results related to dynamic MIS, Maximum Flow, and Maximum Matching. A common trait of all our results is that despite improving state of the art upper bounds or matching state of the art lower bounds, they are surprisingly simple and are analysed using simple amortization arguments. Further, they use no complicated data structures or black box algorithms for their implementation.