Superposition-Assisted Stochastic Optimization for Hawkes Processes

We consider the learning of multi-agent Hawkes processes, a model containing multiple Hawkes processes with shared endogenous impact functions and different exogenous intensities. In the framework of stochastic maximum likelihood estimation, we explore the associated risk bound. Further, we consider the superposition of Hawkes processes within the model, and demonstrate that under certain conditions such an operation is beneficial for tightening the risk bound. Accordingly, we propose a stochastic optimization algorithm assisted with a diversity-driven superposition strategy, achieving better learning results with improved convergence properties. The effectiveness of the proposed method is verified on synthetic data, and its potential to solve the cold-start problem of sequential recommendation systems is demonstrated on real-world data.

[1]  Jure Leskovec,et al.  SEISMIC: A Self-Exciting Point Process Model for Predicting Tweet Popularity , 2015, KDD.

[2]  Ward Whitt,et al.  Characterizing Superposition Arrival Processes in Packet Multiplexers for Voice and Data , 1986, IEEE J. Sel. Areas Commun..

[3]  E. Bacry,et al.  Non-parametric kernel estimation for symmetric Hawkes processes. Application to high frequency financial data , 2011, 1112.1838.

[4]  Aila Särkkä,et al.  Classification of points in superpositions of Strauss and Poisson processes , 2015 .

[5]  George Karypis,et al.  SLIM: Sparse Linear Methods for Top-N Recommender Systems , 2011, 2011 IEEE 11th International Conference on Data Mining.

[6]  Le Song,et al.  Estimating Diffusion Networks: Recovery Conditions, Sample Complexity and Soft-thresholding Algorithm , 2016, J. Mach. Learn. Res..

[7]  Walter L. Smith,et al.  ON THE SUPERPOSITION OF RENEWAL PROCESSES , 1954 .

[8]  Adrian E. Raftery,et al.  Classification of Mixtures of Spatial Point Processes via Partial Bayes Factors , 2005 .

[9]  Le Song,et al.  Coevolutionary Latent Feature Processes for Continuous-Time User-Item Interactions , 2016, NIPS.

[10]  Lars Schmidt-Thieme,et al.  Factorizing personalized Markov chains for next-basket recommendation , 2010, WWW '10.

[11]  Le Song,et al.  Time-Sensitive Recommendation From Recurrent User Activities , 2015, NIPS.

[12]  Erik A. Lewis,et al.  RESEARCH ARTICLE A Nonparametric EM algorithm for Multiscale Hawkes Processes , 2011 .

[13]  Dimitri P. Bertsekas,et al.  Convex Optimization Algorithms , 2015 .

[14]  Jesper Møller,et al.  Transforming Spatial Point Processes into Poisson Processes Using Random Superposition , 2012, Advances in Applied Probability.

[15]  Bamshad Mobasher,et al.  CSLIM: contextual SLIM recommendation algorithms , 2014, RecSys '14.

[16]  Polina Golland,et al.  Sampling from Determinantal Point Processes for Scalable Manifold Learning , 2015, IPMI.

[17]  Evangelia Christakopoulou,et al.  Local Item-Item Models For Top-N Recommendation , 2016, RecSys.

[18]  Aila Särkkä,et al.  Variational Bayes approach for classification of points in superpositions of point processes , 2015 .

[19]  Rong Jin,et al.  Empirical Risk Minimization for Stochastic Convex Optimization: $O(1/n)$- and $O(1/n^2)$-type of Risk Bounds , 2017, COLT.

[20]  Jesper Møller,et al.  Approximate Simulation of Hawkes Processes , 2006 .

[21]  E. Bacry,et al.  A generalization error bound for sparse and low-rank multivariate Hawkes processes , 2015 .

[22]  Hongyuan Zha,et al.  Learning Granger Causality for Hawkes Processes , 2016, ICML.

[23]  Stephen J. Hardiman,et al.  Critical reflexivity in financial markets: a Hawkes process analysis , 2013, 1302.1405.

[24]  E. Çinlar,et al.  On the Superposition of Point Processes , 1968 .

[25]  Le Song,et al.  Learning Social Infectivity in Sparse Low-rank Networks Using Multi-dimensional Hawkes Processes , 2013, AISTATS.

[26]  A. Hawkes.  Point Spectra of Some Mutually Exciting Point Processes , 1971 .

[27]  Susan L. Albin,et al.  Approximating a Point Process by a Renewal Process, II: Superposition Arrival Processes to Queues , 1984, Oper. Res..

[28]  Lars Schmidt-Thieme,et al.  BPR: Bayesian Personalized Ranking from Implicit Feedback , 2009, UAI.

[29]  Niao He,et al.  Online Learning for Multivariate Hawkes Processes , 2017, NIPS.

[30]  Hongyuan Zha,et al.  Learning Hawkes Processes from Short Doubly-Censored Event Sequences , 2017, ICML.

[31]  R. Dahlhaus,et al.  Graphical Modeling for Multivariate Hawkes Processes with Nonparametric Link Functions , 2016, 1605.06759.

[32]  Shuang-Hong Yang,et al.  Mixture of Mutually Exciting Processes for Viral Diffusion , 2013, ICML.

[33]  Xu Chen,et al.  Benefits from Superposed Hawkes Processes , 2017, AISTATS.

[34]  Wenjun Zhang,et al.  Multi-Task Multi-Dimensional Hawkes Processes for Modeling Event Sequences , 2015, IJCAI.

[35]  Le Song,et al.  Shaping Social Activity by Incentivizing Users , 2014, NIPS.

[36]  Le Song,et al.  Isotonic Hawkes Processes , 2016, ICML.

[37]  Lingjiong Zhu,et al.  Ruin Probabilities for Risk Processes with Non-Stationary Arrivals and Subexponential Claims , 2013 .

[38]  Rebecca Willett,et al.  Tracking Dynamic Point Processes on Networks , 2014, IEEE Transactions on Information Theory.

[39]  Emmanuel Bacry,et al.  Uncovering Causality from Multivariate Hawkes Integrated Cumulants , 2016, ICML.

[40]  E. Çinlar.  On the superposition of m-dimensional point processes , 1968, Journal of Applied Probability.

[41]  François Roueff,et al.  Locally stationary Hawkes processes , 2016 .

[42]  Esko Valkeila,et al.  An Introduction to the Theory of Point Processes, Volume II: General Theory and Structure, 2nd Edition by Daryl J. Daley, David Vere‐Jones , 2008 .

[43]  Ulrike Goldschmidt,et al.  An Introduction To The Theory Of Point Processes , 2016 .

[44]  Emmanuel Bacry,et al.  Second order statistics characterization of Hawkes processes and non-parametric estimation , 2014, 1401.0903.

[45]  Julian J. McAuley,et al.  Ups and Downs: Modeling the Visual Evolution of Fashion Trends with One-Class Collaborative Filtering , 2016, WWW.