Fast and Flexible Temporal Point Processes with Triangular Maps
Temporal point process (TPP) models combined with recurrent neural networks provide a powerful framework for modeling continuous-time event data. While such models are flexible, they are inherently sequential and therefore cannot benefit from the parallelism of modern hardware. By exploiting the rec...
Saved in:
Main Authors | , , , |
---|---|
Format | Journal Article |
Language | English |
Published |
22.06.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Temporal point process (TPP) models combined with recurrent neural networks
provide a powerful framework for modeling continuous-time event data. While
such models are flexible, they are inherently sequential and therefore cannot
benefit from the parallelism of modern hardware. By exploiting the recent
developments in the field of normalizing flows, we design TriTPP -- a new class
of non-recurrent TPP models, where both sampling and likelihood computation can
be done in parallel. TriTPP matches the flexibility of RNN-based methods but
permits orders of magnitude faster sampling. This enables us to use the new
model for variational inference in continuous-time discrete-state systems. We
demonstrate the advantages of the proposed framework on synthetic and
real-world datasets. |
---|---|
DOI: | 10.48550/arxiv.2006.12631 |