1: \begin{abstract}

2: We study convergence of the following discrete-time non-linear dynamical system:

3: $n$ agents are located in $\R^d$ and at every time step,

4: each moves synchronously to the average location of all agents within

5: a unit distance of it. This popularly studied system was introduced by

6: Krause to model the dynamics of opinion formation and is often

7: referred to as the {\em Hegselmann-Krause model}. We prove the first

8: polynomial time bound for the convergence of this system in arbitrary

9: dimensions. This improves on the bound of $n^{O(n)}$ resulting from 

10: a more general theorem of Chazelle \cite{Chazelle11}. Also, we show a

11: quadratic lower bound and improve the upper bound for one-dimensional

12: systems to $O(n^3)$. 

13: \end{abstract}