Wasserstein Learning of Deep Generative Point Process Models

Point processes are becoming very popular in modeling asynchronous sequential data due to their sound mathematical foundation and strength in modeling a variety of real-world phenomena. Currently, they are often characterized via intensity function which limits model's expressiveness due to unrealistic assumptions on its parametric form used in practice. Furthermore, they are learned via maximum likelihood approach which is prone to failure in multi-modal distributions of sequences. In this paper, we propose an intensity-free approach for point processes modeling that transforms nuisance processes to a target one. Furthermore, we train the model using a likelihood-free leveraging Wasserstein distance between point processes. Experiments on various synthetic and real-world data substantiate the superiority of the proposed point process model over conventional ones.

[1]  Yoshua Bengio,et al.  Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Olof Mogren,et al.  C-RNN-GAN: Continuous recurrent neural networks with adversarial training , 2016, ArXiv.

[3]  E. Bacry,et al.  Hawkes Processes in Finance , 2015, 1502.04592.

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

[5]  Utkarsh Upadhyay,et al.  Recurrent Marked Temporal Point Processes: Embedding Event History to Vector , 2016, KDD.

[6]  Jason Eisner,et al.  The Neural Hawkes Process: A Neurally Self-Modulating Multivariate Point Process , 2016, NIPS.

[7]  Yosihiko Ogata,et al.  On Lewis' simulation method for point processes , 1981, IEEE Trans. Inf. Theory.

[8]  Vinayak A. Rao,et al.  A Multitask Point Process Predictive Model , 2015, ICML.

[9]  O. Aalen,et al.  Survival and Event History Analysis: A Process Point of View , 2008 .

[10]  Ferenc Huszar,et al.  How (not) to Train your Generative Model: Scheduled Sampling, Likelihood, Adversary? , 2015, ArXiv.

[11]  Marco Cuturi,et al.  Soft-DTW: a Differentiable Loss Function for Time-Series , 2017, ICML.

[12]  Scott W. Linderman,et al.  Discovering Latent Network Structure in Point Process Data , 2014, ICML.

[13]  Aaron C. Courville,et al.  Improved Training of Wasserstein GANs , 2017, NIPS.

[14]  Aihua Xia,et al.  A new metric between distributions of point processes , 2007, Advances in Applied Probability.

[15]  Amitabha Mukerjee,et al.  Contextual RNN-GANs for Abstract Reasoning Diagram Generation , 2016, AAAI.

[16]  Hongyuan Zha,et al.  Modeling the Intensity Function of Point Process Via Recurrent Neural Networks , 2017, AAAI.

[17]  Andrew B. Whinston,et al.  Path to Purchase: A Mutually Exciting Point Process Model for Online Advertising and Conversion , 2012, Manag. Sci..

[18]  L. Decreusefond,et al.  Functional Poisson approximation in Kantorovich-Rubinstein distance with applications to U-statistics and stochastic geometry , 2014, 1406.5484.

[19]  Le Song,et al.  Joint Modeling of Event Sequence and Time Series with Attentional Twin Recurrent Neural Networks , 2017, ArXiv.

[20]  Matthias Bethge,et al.  A note on the evaluation of generative models , 2015, ICLR.

[21]  Shuang Li,et al.  COEVOLVE: A Joint Point Process Model for Information Diffusion and Network Co-evolution , 2015, NIPS.

[22]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[23]  Léon Bottou,et al.  Towards Principled Methods for Training Generative Adversarial Networks , 2017, ICLR.

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

[25]  Ian J. Goodfellow,et al.  NIPS 2016 Tutorial: Generative Adversarial Networks , 2016, ArXiv.

[26]  Hongyuan Zha,et al.  A new Mallows distance based metric for comparing clusterings , 2005, ICML '05.

[27]  Le Song,et al.  Multistage Campaigning in Social Networks , 2016, NIPS.

[28]  A. Hawkes.  Spectra of some self-exciting and mutually exciting point processes , 1971 .

[29]  V. Isham,et al.  A self-correcting point process , 1979 .

[30]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.