Automatic Ensemble Learning for Online Influence Maximization

We consider the problem of selecting a seed set to maximize the expected number of influenced nodes in the social network, referred to as the \textit{influence maximization} (IM) problem. We assume that the topology of the social network is prescribed while the influence probabilities among edges are unknown. In order to learn the influence probabilities and simultaneously maximize the influence spread, we consider the tradeoff between exploiting the current estimation of the influence probabilities to ensure certain influence spread and exploring more nodes to learn better about the influence probabilities. The exploitation-exploration trade-off is the core issue in the multi-armed bandit (MAB) problem. If we regard the influence spread as the reward, then the IM problem could be reduced to the combinatorial multi-armed bandits. At each round, the learner selects a limited number of seed nodes in the social network, then the influence spreads over the network according to the real influence probabilities. The learner could observe the activation status of the edge if and only if its start node is influenced, which is referred to as the edge-level semi-bandit feedback. Two classical bandit algorithms including Thompson Sampling and Epsilon Greedy are used to solve this combinatorial problem. To ensure the robustness of these two algorithms, we use an automatic ensemble learning strategy, which combines the exploration strategy with exploitation strategy. The ensemble algorithm is self-adaptive regarding that the probability of each algorithm could be adjusted based on the historical performance of the algorithm. Experimental evaluation illustrates the effectiveness of the automatically adjusted hybridization of exploration algorithm with exploitation algorithm.

[1]  Wei Chen,et al.  Efficient influence maximization in social networks , 2009, KDD.

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

[3]  Dacheng Tao,et al.  Networked bandits with disjoint linear payoffs , 2014, KDD.

[4]  John Shawe-Taylor,et al.  PAC-Bayes-Bernstein Inequality for Martingales and its Application to Multiarmed Bandits , 2011, ICML On-line Trading of Exploration and Exploitation.

[5]  Benjamin Van Roy,et al.  An Information-Theoretic Analysis of Thompson Sampling , 2014, J. Mach. Learn. Res..

[6]  Shipra Agrawal,et al.  Thompson Sampling for Contextual Bandits with Linear Payoffs , 2012, ICML.

[7]  Alessandro Lazaric,et al.  Linear Thompson Sampling Revisited , 2016, AISTATS.

[8]  Christos Dimitrakakis,et al.  Thompson Sampling for Stochastic Bandits with Graph Feedback , 2017, AAAI.

[9]  Zheng Wen,et al.  Online Influence Maximization under Independent Cascade Model with Semi-Bandit Feedback , 2016, NIPS.

[10]  Andreas Krause,et al.  Cost-effective outbreak detection in networks , 2007, KDD '07.

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

[12]  Michal Valko,et al.  Revealing Graph Bandits for Maximizing Local Influence , 2016, AISTATS.

[13]  Lihong Li,et al.  An Empirical Evaluation of Thompson Sampling , 2011, NIPS.

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

[15]  Xiaokui Xiao,et al.  Influence maximization: near-optimal time complexity meets practical efficiency , 2014, SIGMOD Conference.

[16]  Xiaokui Xiao,et al.  Influence Maximization in Near-Linear Time: A Martingale Approach , 2015, SIGMOD Conference.

[17]  Jinha Kim,et al.  Scalable and parallelizable processing of influence maximization for large-scale social networks? , 2013, 2013 IEEE 29th International Conference on Data Engineering (ICDE).

[18]  Laks V. S. Lakshmanan,et al.  CELF++: optimizing the greedy algorithm for influence maximization in social networks , 2011, WWW.

[19]  Reynold Cheng,et al.  Online Influence Maximization , 2015, KDD.

[20]  Yajun Wang,et al.  Combinatorial Multi-Armed Bandit and Its Extension to Probabilistically Triggered Arms , 2014, J. Mach. Learn. Res..

[21]  Kyomin Jung,et al.  IRIE: Scalable and Robust Influence Maximization in Social Networks , 2011, 2012 IEEE 12th International Conference on Data Mining.

[22]  Laks V. S. Lakshmanan,et al.  A Data-Based Approach to Social Influence Maximization , 2011, Proc. VLDB Endow..

[23]  Laks V. S. Lakshmanan,et al.  Diffusion Independent Semi-Bandit Influence Maximization , 2017, ArXiv.

[24]  Laks V. S. Lakshmanan,et al.  Model-Independent Online Learning for Influence Maximization , 2017, ICML.

[25]  Laks V. S. Lakshmanan,et al.  Influence Maximization with Bandits , 2015, ArXiv.

[26]  Rémi Munos,et al.  Thompson Sampling: An Asymptotically Optimal Finite-Time Analysis , 2012, ALT.

[27]  Michal Valko Bandits on graphs and structures , 2016 .

[28]  Rémi Munos,et al.  Thompson Sampling for 1-Dimensional Exponential Family Bandits , 2013, NIPS.

[29]  Csaba Szepesvári,et al.  Improved Algorithms for Linear Stochastic Bandits , 2011, NIPS.

[30]  Zheng Wen,et al.  Influence Maximization with Semi-Bandit Feedback , 2016, ArXiv.

[31]  Wei Chen,et al.  Scalable influence maximization for prevalent viral marketing in large-scale social networks , 2010, KDD.

[32]  Shipra Agrawal,et al.  Analysis of Thompson Sampling for the Multi-armed Bandit Problem , 2011, COLT.

[33]  Cem Tekin,et al.  Online Contextual Influence Maximization in social networks , 2016, 2016 54th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[34]  Wei Chen,et al.  Combinatorial Multi-Armed Bandit: General Framework and Applications , 2013, ICML.

[35]  Steven L. Scott,et al.  A modern Bayesian look at the multi-armed bandit , 2010 .