Wasserstein Learning of Deep Generative Point Process Models
arXiv:1705.08051
Abstract
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.
References in corpus (6)
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- Improved Training of Wasserstein GANs
- NIPS 2016 Tutorial: Generative Adversarial Networks
- A note on the evaluation of generative models
- C-RNN-GAN: Continuous recurrent neural networks with adversarial training
- How (not) to Train your Generative Model: Scheduled Sampling, Likelihood, Adversary?
Cited by in corpus (6)
- Deep Learning for Time Series Forecasting: Tutorial and Literature Survey
- Synthetic Observational Health Data with GANs: from slow adoption to a boom in medical research and ultimately digital twins?
- Long Horizon Forecasting With Temporal Point Processes
- Modeling Event Propagation via Graph Biased Temporal Point Process
- THAP: A Matlab Toolkit for Learning with Hawkes Processes
- Hawkes Processes on Graphons