Saved in:
Bibliographic Details
Main Authors: Garza, Azul, Challu, Cristian, Mergenthaler-Canseco, Max
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2310.03589
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910460417146880
author Garza, Azul
Challu, Cristian
Mergenthaler-Canseco, Max
author_facet Garza, Azul
Challu, Cristian
Mergenthaler-Canseco, Max
contents In this paper, we introduce TimeGPT, the first foundation model for time series, capable of generating accurate predictions for diverse datasets not seen during training. We evaluate our pre-trained model against established statistical, machine learning, and deep learning methods, demonstrating that TimeGPT zero-shot inference excels in performance, efficiency, and simplicity. Our study provides compelling evidence that insights from other domains of artificial intelligence can be effectively applied to time series analysis. We conclude that large-scale time series models offer an exciting opportunity to democratize access to precise predictions and reduce uncertainty by leveraging the capabilities of contemporary advancements in deep learning.
format Preprint
id arxiv_https___arxiv_org_abs_2310_03589
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TimeGPT-1
Garza, Azul
Challu, Cristian
Mergenthaler-Canseco, Max
Machine Learning
Applications
In this paper, we introduce TimeGPT, the first foundation model for time series, capable of generating accurate predictions for diverse datasets not seen during training. We evaluate our pre-trained model against established statistical, machine learning, and deep learning methods, demonstrating that TimeGPT zero-shot inference excels in performance, efficiency, and simplicity. Our study provides compelling evidence that insights from other domains of artificial intelligence can be effectively applied to time series analysis. We conclude that large-scale time series models offer an exciting opportunity to democratize access to precise predictions and reduce uncertainty by leveraging the capabilities of contemporary advancements in deep learning.
title TimeGPT-1
topic Machine Learning
Applications
url https://arxiv.org/abs/2310.03589