Add and Thin: Diffusion for Temporal Point Processes

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Lüdke, David, Biloš, Marin, Shchur, Oleksandr, Lienen, Marten, Günnemann, Stephan
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866929248482099200
author Lüdke, David
Biloš, Marin
Shchur, Oleksandr
Lienen, Marten
Günnemann, Stephan
author_facet Lüdke, David
Biloš, Marin
Shchur, Oleksandr
Lienen, Marten
Günnemann, Stephan
contents Autoregressive neural networks within the temporal point process (TPP) framework have become the standard for modeling continuous-time event data. Even though these models can expressively capture event sequences in a one-step-ahead fashion, they are inherently limited for long-term forecasting applications due to the accumulation of errors caused by their sequential nature. To overcome these limitations, we derive ADD-THIN, a principled probabilistic denoising diffusion model for TPPs that operates on entire event sequences. Unlike existing diffusion approaches, ADD-THIN naturally handles data with discrete and continuous components. In experiments on synthetic and real-world datasets, our model matches the state-of-the-art TPP models in density estimation and strongly outperforms them in forecasting.
format Preprint
id arxiv_https___arxiv_org_abs_2311_01139
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Add and Thin: Diffusion for Temporal Point Processes
Lüdke, David
Biloš, Marin
Shchur, Oleksandr
Lienen, Marten
Günnemann, Stephan
Machine Learning
Autoregressive neural networks within the temporal point process (TPP) framework have become the standard for modeling continuous-time event data. Even though these models can expressively capture event sequences in a one-step-ahead fashion, they are inherently limited for long-term forecasting applications due to the accumulation of errors caused by their sequential nature. To overcome these limitations, we derive ADD-THIN, a principled probabilistic denoising diffusion model for TPPs that operates on entire event sequences. Unlike existing diffusion approaches, ADD-THIN naturally handles data with discrete and continuous components. In experiments on synthetic and real-world datasets, our model matches the state-of-the-art TPP models in density estimation and strongly outperforms them in forecasting.
title Add and Thin: Diffusion for Temporal Point Processes
topic Machine Learning
url https://arxiv.org/abs/2311.01139