The Importance of Communities for Learning to Influence

We consider the canonical problem of influence maximization in social networks. Since the seminal work of Kempe, Kleinberg, and Tardos, there have been two largely disjoint efforts on this problem. The first studies the problem associated with learning the parameters of the generative influence model. The second focuses on the algorithmic challenge of identifying a set of influencers, assuming the parameters of the generative model are known. Recent results on learning and optimization imply that in general, if the generative model is not known but rather learned from training data, no algorithm can yield a constant factor approximation guarantee using polynomially-many samples, drawn from any distribution. In this paper, we design a simple heuristic that overcomes this negative result in practice by leveraging the strong community structure of social networks. Although in general the approximation guarantee of our algorithm is necessarily unbounded, we show that this algorithm performs well experimentally. To justify its performance, we prove our algorithm obtains a constant factor approximation guarantee on graphs generated through the stochastic block model, traditionally used to model networks with community structure.

[1]  Christian Borgs,et al.  Maximizing Social Influence in Nearly Optimal Time , 2012, SODA.

[2]  Éva Tardos,et al.  Influential Nodes in a Diffusion Model for Social Networks , 2005, ICALP.

[3]  Le Song,et al.  Learning Networks of Heterogeneous Influence , 2012, NIPS.

[4]  Eric Balkanski,et al.  The limitations of optimization from samples , 2015, STOC.

[5]  Laks V. S. Lakshmanan,et al.  Learning influence probabilities in social networks , 2010, WSDM '10.

[6]  Wei Chen,et al.  Robust Influence Maximization , 2016, KDD.

[7]  Le Song,et al.  Estimating Diffusion Network Structures: Recovery Conditions, Sample Complexity & Soft-thresholding Algorithm , 2014, ICML.

[8]  Matthew Richardson,et al.  Mining the network value of customers , 2001, KDD '01.

[9]  Eric Balkanski,et al.  The Sample Complexity of Optimizing a Convex Function , 2017, COLT.

[10]  Niloy Ganguly,et al.  Learning a Linear Influence Model from Transient Opinion Dynamics , 2014, CIKM.

[11]  Eric Balkanski,et al.  The Power of Optimization from Samples , 2016, NIPS.

[12]  Sujay Sanghavi,et al.  Learning the graph of epidemic cascades , 2012, SIGMETRICS '12.

[13]  Jon M. Kleinberg,et al.  Reconstructing Patterns of Information Diffusion from Incomplete Observations , 2011, NIPS.

[14]  Yan Liu,et al.  Learning Influence Functions from Incomplete Observations , 2016, NIPS.

[15]  Luis E. Ortiz,et al.  Learning the structure and parameters of large-population graphical games from behavioral data , 2012, J. Mach. Learn. Res..

[16]  Yaron Singer,et al.  Scalable Methods for Adaptively Seeding a Social Network , 2015, WWW.

[17]  Jon M. Kleinberg,et al.  The link-prediction problem for social networks , 2007, J. Assoc. Inf. Sci. Technol..

[18]  Alessandro Panconesi,et al.  Trace complexity of network inference , 2013, KDD.

[19]  P. Erdos,et al.  On the evolution of random graphs , 1984 .

[20]  Jure Leskovec,et al.  Inferring networks of diffusion and influence , 2010, KDD.

[21]  Yaron Singer,et al.  Maximization of Approximately Submodular Functions , 2016, NIPS.

[22]  Elchanan Mossel,et al.  Submodularity of Influence in Social Networks: From Local to Global , 2010, SIAM J. Comput..

[23]  Bernhard Schölkopf,et al.  Uncovering the Temporal Dynamics of Diffusion Networks , 2011, ICML.

[24]  Lada A. Adamic,et al.  Tracking information epidemics in blogspace , 2005, The 2005 IEEE/WIC/ACM International Conference on Web Intelligence (WI'05).

[25]  Jure Leskovec,et al.  {SNAP Datasets}: {Stanford} Large Network Dataset Collection , 2014 .

[26]  Tim Roughgarden,et al.  Sketching valuation functions , 2012, SODA.

[27]  Rediet Abebe Can Cascades be Predicted? , 2014 .

[28]  Christos Faloutsos,et al.  Patterns of Cascading Behavior in Large Blog Graphs , 2007, SDM.

[29]  Matthew Richardson,et al.  Mining knowledge-sharing sites for viral marketing , 2002, KDD.

[30]  David C. Parkes,et al.  Learnability of Influence in Networks , 2015, NIPS.

[31]  Jon Kleinberg,et al.  Maximizing the spread of influence through a social network , 2003, KDD '03.

[32]  Grant Schoenebeck,et al.  Don't Be Greedy: Leveraging Community Structure to Find High Quality Seed Sets for Influence Maximization , 2017, WINE.

[33]  Le Song,et al.  Scalable Influence Estimation in Continuous-Time Diffusion Networks , 2013, NIPS.

[34]  Lior Seeman,et al.  Adaptive Seeding in Social Networks , 2013, 2013 IEEE 54th Annual Symposium on Foundations of Computer Science.

[35]  Pravesh Kothari,et al.  Learning Coverage Functions and Private Release of Marginals , 2014, COLT.

[36]  Jon Kleinberg,et al.  The link prediction problem for social networks , 2003, CIKM '03.

[37]  Éva Tardos,et al.  Maximizing the Spread of Influence through a Social Network , 2015, Theory Comput..

[38]  Le Song,et al.  Influence Function Learning in Information Diffusion Networks , 2014, ICML.