1: \begin{abstract}

2:   %\emph{``THIS PAPER IS ELIGIBLE FOR THE STUDENT PAPER AWARD.''}

3:   Many modern applications involve accessing 

4:   and processing graphical

5:   data, i.e. data that is naturally indexed by graphs. Examples come from

6:   internet graphs, social networks, genomics and

7:   proteomics, and other sources. The typically large size of such data motivates seeking efficient ways for its compression and decompression.

8:   The current compression  methods are usually tailored to specific models, or do not provide

9:   theoretical guarantees. 

10:   In  this paper, we introduce a low--complexity lossless compression algorithm for sparse marked graphs, i.e. graphical data indexed by sparse graphs, which is

11:   capable of universally achieving the optimal compression rate in a precisely defined sense. In order to

12:   define  universality, we employ the framework of local weak convergence,

13:   which allows one to make sense of a notion of stochastic processes for sparse graphs.

14:   %Moreover, we use a generalized version of the entropy introduced by Bordenave and Caputo

15:   %to discuss information theoretic optimality.

16:   Moreover, we investigate the performance of

17:   our algorithm through some experimental results on both synthetic and

18:   real--world data. 

19: \end{abstract}

20: