Corpus ID: 202708306

Generative Sequential Stochastic Model for Marked Point Processes

@inproceedings{Sharma2019GenerativeSS,
  title={Generative Sequential Stochastic Model for Marked Point Processes},
  author={Abhishek Sharma and Aritra Ghosh and M. Fiterau},
  year={2019}
}
Temporal point processes are often used to model event data streams in the real world, such as financial transactions, electronic health records, and social networks. While several parametric approaches, such as Poisson and Hawkes processes are used to model temporal dynamics, they often suffer from the curse of model misspecification. More flexible RNN based approaches have been recently proposed to jointly model event time and marker information through the use of a shared hidden… Expand

Figures and Tables from this paper

Latent Embeddings of Point Process Excitations
Event Cartography: Latent Point Process Embeddings

References

SHOWING 1-10 OF 27 REFERENCES
Learning Conditional Generative Models for Temporal Point Processes
Modeling the Intensity Function of Point Process Via Recurrent Neural Networks
Marked Temporal Dynamics Modeling Based on Recurrent Neural Network
A Dirichlet Mixture Model of Hawkes Processes for Event Sequence Clustering
The Neural Hawkes Process: A Neurally Self-Modulating Multivariate Point Process
Recurrent Spatio-Temporal Point Process for Check-in Time Prediction
Wasserstein Learning of Deep Generative Point Process Models
Structured Inference Networks for Nonlinear State Space Models
...
1
2
3
...