A Variational Autoencoder for Neural Temporal Point Processes with Dynamic Latent Graphs

arXiv:2312.16083v2 Announce Type: replace
Abstract: Continuously-observed event occurrences, often exhibit self- and mutually-exciting effects, which can be well modeled using temporal point processes. Beyond that, these event dynamics may also change over time, with certain periodic trends. We propose a novel variational auto-encoder to capture such a mixture of temporal dynamics. More specifically, the whole time interval of the input sequence is partitioned into a set of sub-intervals. The event dynamics are assumed to be stationary within each sub-interval, but …

abstract arxiv auto autoencoder beyond change cs.lg dynamic dynamics effects encoder event graphs novel processes stat.ml temporal trends type